Marcus L Endicott: Best Quora Answers 2014

Marcus L Endicott’s Answers – Quora

See also:

Marcus L Endicott: Best Quora Answers 2011 | Marcus L Endicott: Best Quora Answers 2012 | Marcus L Endicott: Best Quora Answers 2013


Mon, Dec 15, 2014, Continue reading at the source
Mon, Dec 15, 2014, Continue reading at the source
Sat, Dec 13, 2014, Continue reading at the source

How do servers and interfaces handle web notifications instantly?

I’m still struggling to fully understand the overall picture myself.  I believe this is generally referred to as the Real-time web.  I can think of three kinds of technologies that contribute to virtual realtime:

  1. Message oriented middleware
  2. Push technology
  3. Streaming media (aka Streaming data)

Written 27 Sep

= = =

Where do I start if I am interested in learning machine learning in order to make a project on recommender systems?

Check out my quick and dirty webpages:

See also my Quora answers for:

Written 26 Sep

= = =

What is best way to startup a professional augmented reality? What are the best tools?

For AR development toolkits, check the List of augmented reality software on Wikipedia.

Otherwise, my quick and dirty webpages should help get you started:

Written 26 Sep

= = =

How does TripAdviser work?

I’m going to hazard a guess that TripAdvisor and similar websites do in fact have their core data onboard, and do not do realtime web crawling or web scraping.

In general, most such core data originates from:

In the case of TripAdvisor, a large part of their added value is crowd sourced, collective intelligence, in other words human feedback, something like Folksonomy. (Incidentally, this same kind of feedback loop was largely responsible for the initial success of Lonely Planet guidebooks, thanks to their loyal fan base.)

Updated 25 Sep

= = =

Is there an API that returns Siri like responses in text when questions are asked in text?

What are the best dialog system APIs for the Internet of Things?

Found so far:

  • allthingstalk.com
  • api.ai
  • ask-ziggy.com
  • botlibre.com
  • chatbot4u.com
  • mondobot.com
  • pandorabots.com
  • wit.ai

Updated 29 Oct

= = =

What language is Siri written in?

Is anyone working on an open source version of Siri?

Siri is based on a government sponsored (SRI) project called PAL (Personalized Assistant that Learns), and *some* of the PAL Framework components are open source. (Actually, CALO, the Cognitive Agent that Learns and Organizes, was originally developed as part of the PAL project.)

Which topics on Quora are related to the technology behind Apple Siri?

Originally, Siri was built on SRI International CALO (Cognitive Assistant that Learns and Organizes). What is not so widely known is that SPARK (SRI Procedural Agent Realization Kit) forms the heart of CALO’s task execution.

Written 22 Sep

= = =

Would all the data from Quora be helpful for making A.I. like Watson be more humanlike in its conversation? If so, how? If not, why?

=> quora.com/search?q=marcus-l-endicott+vicarious

Versions of this question seem to get asked over and over again on Quora, above. Generally, conversational AI and in particular question answering systems are built on top of question/answer pairs, and machine learning can be applied to question/answer pairs. The biggest stumbling block at the moment is the conspicuous absence of a publicly available Quora API. Quora co-founder Adam D’Angelo may already be at work on this through his stake in Vicarious (company).

Written 19 Sep

= = =

What are good alternatives to NLTK?

Wikipedia contains an extensive listing of Natural language processing toolkits, under Outline of natural language processing.

See also my recent answer to:

In light of the updated comment, I have presented a metric, see references below:

  • NLTK (Natural Language Toolkit) .. Python (240)
  • OpenNLP .. Java (38)
  • MALLET (MAchine Learning for LanguagE Toolkit) .. Java (34)
  • Stanford NLP .. Java (31)
  • LingPipe .. Java (5)
  • FreeLing .. Ruby (4)
  • TreeTagger .. Ruby (2)

Note, ClearNLP may be added to the above list:

  • ClearNLP .. Java (6)

Written 18 Sep

= = =

How do I go about developing my own chatbot?

=> quora.com/Marcus-L-Endicott/answers

Please read through all my Quora answers above, specifically on and around this topic. I’ll try to address interesting questions under the following Quora topics:

  1. Conversational Agents (Chatbots)
  2. Dialog System (software)
  3. Virtual Assistants (bots)

Written 17 Sep

= = =

What are the different layers of cloud computing?

I generally divide cloud offerings first into two kinds, the commandline cloud (including IaaS and PaaS) and the SaaS cloud. There are of course also hybrid offerings which include both commandline and SaaS components. One example of commandline cloud would be Heroku, and for SaaS cloud would be Salesforce. Though in fairness Salesforce would also include hybrid elements. I define these hybrid clouds as ones that allow you to get under the hood and tweak the code. The SaaS cloud is not necessarily visual, as in GUI; I believe systems offering APIs to also be part of the SaaS cloud and hybrid SaaS cloud.

Written 17 Sep

= = =

Is there any artificial intelligence software or OS? Or is any company working on that project?

This is something I have been looking at for a while. What is needed most is good artificial intelligence middleware, for Artificial intelligence systems integration. Unfortunately, when people speak of AI middleware, they are generally referring game AI middleware, which is not appropriate for real world applications. There are a number of robot operating systems available, such as the popular ROS. And, inter-connectivity of the Internet of Things is literally just around the corner.

Generally when people talk about this kind of thing, especially in the context of movies, they are referring to a bundled AI system. There are two main kinds of bundled AI systems available. There are bundled AI systems for the desktop, and there are expensive, enterprise-grade bundled AI solutions for the web. I generally recommend Guile3D Studio for the typical user, for instance DIY home automation.

See also my quick and dirty webpages:

Written 17 Sep

= = =

Gnip vs Data Sift – which is better in pricing?

You just punched the wrong button…. There are a number of issues here. First of all, Gnip has been acquired by Twitter. I have more experience with Texifter Sifter than with DataSift. I am seriously unhappy with the pricing structure presented to me by Sifter on behalf of DataSift. The reality is that they are charging an arm and a leg, for your tweets and my tweets given freely in good faith, in fact more than the price for gold. Sifter has told me that their hands are tied by their supplier, DataSift. In any case, the bottom line is you must pay $20 per day in addition to whatever quantity is retrieved; so, even when your query is relatively constrained, if you go back say 100 days then instantly add $2,000 to the bill. I say balderdash! (And, that’s when I’m in a good mood.)

Written 16 Sep

= = =

Is it possible to integrate all customer service numbers into Google speech recognition, so we can directly ask questions?

It is not completely clear to me what you are asking…. It seems Skype offers this integration with any number on the web, as long as you have some Skype-out credits. It is much more common to have live “chat” (IM) for customer service on the web.

You may be asking about a kind of Intercom system for your website, using Google Web Speech API. The drawback here is that the Web Speech API is limited to the Google Chrome browser, I believe. Also, as far as I know, the Web Speech API is still undocumented. In any case, I am not aware of any CRM system employing this technique to date.

You may wish to refer to my quick and dirty webpage:

Written 16 Sep

= = =

How can governments help the transition as more and more jobs are automated or replaced by robots?

1) The taxation structure surrounding profiteering through automation and robotic production needs to be re-examined, re-envisioned, and re-vamped.

2) The logical response to shrinking employment would be incrementally shortening the working week, as well as gradually lowering the pensionable retirement age.

3) Socialism will be the only way forward. This means a basic living standard, including basic healthcare and education, guaranteed for everyone in the human family, or at least fellow citizens.

Written 16 Sep

= = =

Is there a free alternative to Cleverbot/Cleverscript?

What are the best dialog system APIs for the Internet of Things?

Found so far:

  • allthingstalk.com
  • api.ai
  • ask-ziggy.com
  • botlibre.com
  • chatbot4u.com
  • mondobot.com
  • pandorabots.com
  • wit.ai

Updated 29 Oct

= = =

How do I make a natural language processing question answering system? What software or toolkits can be used?

I plan to make a Question Answering System for my college using NLP. What could be a good approach towards building this system? Also, how could I structure my database for the same?

Start by reviewing my Quora answers about question answering:

  1. What are some good papers to read about Question Answering?
  2. How do I make a data-driven question answering system from scratch?
  3. Question and answer based search engine using DBpedia?
  4. What are the key challenges in designing/developing an effective question answering system?
  5. How do I generate questions from corpus or comprehensions using NLP concepts?
  6. What percentage of questions would IBM’s Watson be able to answer on Quora?
  7. What is the current status of systems which can answer questions framed from a given text article?
  8. Are there any web sites that integrate social question-answering with conversational bots?
  9. Can Quora Q&A be fed into a robot so that it could answer questions like a human?
  10. What are some creative ideas for Quora’s (future) API?
  11. How does IBM Watson advance question answering in general?
  12. How would you build a robot to answer questions on Quora?
  13. What are some examples of generalizable question-answering methods for specialized domains?
  14. What’s best way to train an chatbot (e.g. alicebot)?
  15. What are some good ways to start on a long AI project?
  16. Would an AI be allowed to answer questions on Quora?
  17. What are the most popular social Q&A sites in France?
  18. Will robots take over Quora?

Updated 15 Sep

= = =

To make a project like Siri or Cortana using artificial intelligence, what links, materials, or blogs would be helpful?

I suggest starting with Quora by reading through the answers under the following topics:

  1. Conversational Agents (Chatbots)
  2. Dialog System (software)
  3. Virtual Assistants (bots)

Written 11 Sep

= = =

What are the basics for artificial intelligence?

Joscha Bach (@Plinz) covers the basics in his 2009 book, Principles of Synthetic Intelligence PSI: An Architecture of Motivated Cognition. Here is a summary in the 2013 video, How to Build a Mind – Artificial Intelligence Reloaded:

For software and business applications, see both the micropsi.com and micropsi-industries.com websites.

See also my quick and dirty webpage:
· MicroPsi Agent Architecture | Meta-Guide.com

11 Sep 2014

= = =

Should Artificial Intelligence be modelled to be kind like humans?

See Kevin Lacker‘s January 2011 answer to my very first Quora question:

Should chat bots have a first and last name, like people do?
No, it is too creepy. It evokes images of a dystopian future where faceless entities control us all, like in Jennifer Government.
Think of cute adorable robots like Wall-E or tweenbots and try to make things like them. Your chat bot will be inept in many ways and if you want people to interact with it regardless then they should perceive it as childlike rather than incompetent.

11 Sep 2014

= = =

Is it possible to do programming in Cloud like we edit documents in Google Docs?

Yes, this type of SaaS is generally referred as a cloud IDE or cloud based IDE, which is short for Integrated Development Environment. In December 2013, ProgrammableWeb listed:
· 12 Cloud-Based IDEs Boost Productivity, ROI

11 Sep 2014

= = =

How is Prolog being used with deep learning in question answering systems?

=> quora.com/search?q=prolog+”ibm-watson”

There has already been a good bit of discussion on Quora about the role of Prolog in IBM Watson, above. More generally, this has been expressed as the role of rule based artificial intelligence versus machine learning in artificial intelligence. The broader field is referred to as automated question answeringor question answering systems, listed in Wikipedia simply as Question answering.

In particular, see the February 2012 Quora answer from IBM Watson researcher James William Murdock IV to:
· How does IBM’s Watson work?

11 Sep 2014

= = =

Is it worth writing your own summarization algorithm or are there any APIs which are good enough?

=> quora.com/search?q=marcus-l-endicott+summarization

See all my Quora answers on summarization, such as:

What is the best tool to summarize a text document?
IMHO, there is no best tool to summarize a text document….

However, Orlando Kalossakas has recently recommended Stremor Automated Summary and Abstract Generator API at Mashape, which also lists:

mashape.com/explore?query=summarizer
· Summarizer Tool API
· PithPicker Engine API
· Text Summarization API

9 Sep 2014

= = =

Should I learn Bayesian networks before deep belief networks?

You should definitely learn Bayesian networks (aka belief networks) beforeDeep belief networks. ;^)

See also my quick and dirty webpages:
· 100 Best Bayesian Network Videos | Meta-Guide.com
· 100 Best Bayesian Tutorial Videos | Meta-Guide.com
· 100 Best Naive Bayes Videos | Meta-Guide.com

9 Sep 2014

= = =

What is the best program with the best AI available to interact with?

“Best program” and “best AI” are two different beasts. Most basically, there are three general categories of “products” available:
1. free and open source packages
2. for-profit, commercial products
3. professional, enterprise-grade solutions

There are plenty of desktop products available, fewer end to end mobile products, and fewer still bundled server products; however, there are a number of managed enterprise-grade solutions available for a price. Typically, these products are either heavy on the interface side, in other words theembodiment, or heavy on the AI side, in other words intelligence. I have not seen one yet that is both really good on the interface side, and really good on the intelligence side.

In terms of intelligence, it basically comes in two types, managed or autonomous. Managed intelligence means that you have complete control over what it says, and autonomous means you don’t really know what it will say or where the answers are coming from. So far, autonomous intelligence is not appropriate for business purposes, such as CRM and Social CRM.

I would say Guile3D Studio is probably the best all around product at the moment for the typical user, for instance DIY home automation; however, this is primarily a desktop product, with a separate mobile product in the pipeline, and no proper server product yet.

See my recent Quora answers to:
· Are there any software tools or APIs that will allow you to create a 3D talking head that will generate correct lip movements when given a text stream?
· What are some good virtual assistant bots?

See also my quick and dirty webpage:
· 100 Best Guile3D Denise Videos | Meta-Guide.com

9 Sep 2014

= = =

What are the best resources for getting started with partially observable Markov decision processes (POMDP)?

Try starting with my quick and dirty webpage:
· Best Markov Decision Process Videos | Meta-Guide.com

7 Sep 2014

= = =

Speech Recognition: How do I make a robot which starts by hearing the command “on” and stops with “off”?

It really depends on what your hardware platform is, and what operating system you’re using…. However, there has been a lot of discussion on Quora about open source speech recognition:
· quora.com/search?q=open-source+speech-recognition

See also Wikipedia:
· List of speech recognition software

Edit: If it’s a hardware solution you’re after, I would simply use an Android phone for the backbone. For more ideas, check my quick and dirty webpages:
· 100 Best Arduino Voice Control Videos | Meta-Guide.com
· 100 Best RaspberryPi Voice Control Videos | Meta-Guide.com
· 100 Best Robot Voice Control Videos | Meta-Guide.com

7 Sep 2014

= = =

Is there any comprehensive book which starts with basics of artificial neural networks and machine learning and covers deep learning as well?

I often find videos more digestible than books, especially for dense subject matter. Take a look at my Meta Guide Videography, and in particular the following quick and dirty webpages:
· 100 Best Deep Learning Videos | Meta-Guide.com
· 100 Best Machine Learning Videos | Meta-Guide.com
· 100 Best Neural Network Videos | Meta-Guide.com

5 Sep 2014

= = =

Is there an iOS app that can read out loud to me (text to speech)?

Sure, there are many. I particularly like SoundGecko.

5 Sep 2014

= = =

Who are the top players in the Indian market in the field of home automation?

Ten Indian startups operating in the smart & wearable devices, IoT space
Founded in 2012 by Srihari T and Hari Narayanan, Pluggx offers a smart device that allows users to control their home appliances and electronics via smartphones. The startup claims that unlike traditional smart home automation systems, Pluggx allows users to control their home and also learns from their habits, lifestyle and schedule, to automatically suggest ways to save money.

Who’s powering the home automation systems in Bangalore’s swankiest apartments?
Silvan Labs is 50 member strong with a couple of offices in Bangalore – an office which houses the employees and an experience center where clients can come and see the solutions in action. With 5 strong people at the helm and the likes of Sharad Sharma on the board, Silvan Labs is confident of their hold on the Bangalore market and are looking to quickly expand to Mumbai and Delhi. Silvan Labs is also looking to raise an investment round to boost their expansion plans.

4 Sep 2014

= = =

What are some examples of ontology in computer science?

For details, see my quick and dirty webpage:
· Best Ontology Tutorial Videos | Meta-Guide.com

4 Sep 2014

= = =

What is the best tool for reposting multiple rss feeds to multiple Twitter accounts?

dlvr.it allows you to post to multiple Twitter accounts from a single dlvr.it account. I use it for my circa 250x Meta-Guide.com “country bots”, one for every country on Earth:
· Meta Guide Country Bots By Alphabet | Meta-Guide.com
· Meta Guide Country Bots By Continent | Meta-Guide.com
· Meta Guide Country Bots By Followers | Meta-Guide.com
· Meta Guide Country Bots By Last Tweet | Meta-Guide.com
· Meta Guide Country Bots By Population | Meta-Guide.com
· Meta Guide Country Bots Out Of Timeline | Meta-Guide.com

4 Sep 2014

= = =

What are the best web parser APIs?

Do you really want a web parser, or web scraper? See my recent Quora answer to: What is the difference between crawling, parsing and scraping?

See also:
· kimono : Turn websites into structured APIs from your browser in seconds

4 Sep 2014

= = =

What is the alternative to machine learning?

cortical.io presents semantic fingerprints and semantic maps as an alternative to machine learning in How Mimicking Brain Function is Revolutionising NLP.

4 Sep 2014

= = =

What is the best text to speech software?

=> quora.com/search?q=text-to-speech

There has been a tremendous amount of discussion on Quora about this already. Note, that Wikipedia has defaulted to the terminology of Speech synthesis, instead of TTS. For that matter, Wikipedia contains a Comparison of speech synthesizers. BTW, Nickolay Shmyrev is a good person to follow on Quora for speech technology answers.

See also my quick and dirty webpages:
· 100 Best Speech Synthesis Videos | Meta-Guide.com
· 100 Best Speech Synthesizer Videos | Meta-Guide.com
· 100 Best Text-To-Speech Tutorial Videos | Meta-Guide.com

4 Sep 2014

= = =

How does the brain create abstractions?

1) I believe the brain to be an imperfect inversion of the Universe itself, something like a 3D holographic mirror – as above, so below….

2) I believe dreams and metaphor will eventually be proven central to the abstraction process.

3) I believe this imagaic abstraction process operates like a “clutch” mechanism, in a car for instance.

4) I believe the key to the abstraction process lies in the mechanism which the brain uses to convert words into images, and images into words.

See my Quora answers related to metaphor:
· quora.com/search?q=marcus-l-endicott+metaphor

4 Sep 2014

= = =

What is the difference between crawling, parsing and scraping?

As per usual, Wikipedia provides pretty good definitions for Web crawler,Web scraping, and Parsing. Scraping usually comes before parsing, though scraping and parsing may be used interchangeably. Crawling and scraping are also often used interchangeably.
· Crawling is essentially following links, both internal and external.
· Scraping is the act of extraction, for instance from crawling.
· Parsing is basically breaking it down into pieces, constituent parts, or segments.

See also my quick and dirty webpages:
· 100 Best Web Crawler Videos | Meta-Guide.com
· 100 Best Web Scraping Videos | Meta-Guide.com
· 100 Best Parsing Tutorial Videos | Meta-Guide.com

3 Sep 2014

= = =

What tools/framework can I use in order for my system to extract information from an input text and store the extracted information on a database and be retrieved for later processing?

The systematic study of text is known as Corpus linguistics. Generally, a corpus can be parsed into and represented as XML, which can then be processed using regular expressions. Wikipedia contains an extensive listing of Natural language processing toolkits, under Outline of natural language processing.

2 Sep 2014

= = =

What is the next big thing in recommendation systems?

IMHO, it’s not about recommendation systems per se, it’s about the technology underlying recommender systems, predictive analytics. For example, supposedly predictive analytics were the core advancement of IBM Watson. Do we generally consider Watson to be a recommender system? No, however predictive analytics recommended the questions for Jeopardy clues. Predictive analytics may eventually play a role in every little thing in the Internet of Things, aka Internet of Everything.

1 Sep 2014

= = =

Why can’t IBM Watson rewrite and improve its own code?

The simple answer is that no one has programmed it to do that. However, just look to the new IBM Watson solutions in the medical field. They are not providing answers per se, but rather graphically representing flows pointing in the best direction. Conceivably, programmers could use this same technique topoint them in the right direction for modifying code.

See my quick and dirty webpage:
· 100 Best IBM Watson Medical Videos | Meta-Guide.com

1 Sep 2014

= = =

What is different about the Salesforce which makes it the biggest player in SaaS?

Salesforce empowers non-programmers to create their own solutions. Salesforce is an umbrella system that integrates other systems. I don’t know of another system that integrates more systems, more easily than Salesforce. My main critique of Salesforce is that it is too rigidly focused on conventional business process, and does not allow enough leeway for the Internet of Things, much less for experimental AI….

See also my quick and dirty webpages:
· 100 Best Salesforce Integration Videos | Meta-Guide.com
· 100 Best Salesforce SocialCRM Videos | Meta-Guide.com
· Best Salesforce Agents | Meta-Guide.com

1 Sep 2014

= = =

What are some good research publications on (or resources to learn about) abstraction based document summarization?

Abstraction-based summarization is more commonly known as abstractive summarization. Truely abstractive summarization may include:
1. sentence reduction
2. sentence compression
3. sentence reformulation

rali.iro.umontreal.ca/rali/?q=en/Automatic summarization
Researchers at RALI, under the direction of Guy Lapalme, have been working in automatic summarization for many years.
Since 2002, RALI has systematically participated in competitions ofDocument Understanding Conferences (DUC), more recently Text Analysis Conference (TAC).


SumUM was developed by Horacio Saggion in his Ph.D. thesis (1997-2000)….
SumUM produces a summary in two steps: first an indicative summary, which identifies important subjects of a document; then an informative summary is produced which elaborates a few subjects selected by the user.

For more references, see my quick and dirty webpage:
· Abstractive Summarization 2013 | Meta-Guide.com

1 Sep 2014

= = =

Can you recommend an automation module (PaaS/SaaS/IaaS)?

I’ve tested the new IBM Bluemix, with the integrated Node-RED GUI, as described in Create a twitter bot on Bluemix using Node-RED. Yahoo Pipes it’s not, but does seem to work okay.

See also my quick and dirty webpage:
· 100 Best IBM Bluemix Videos | Meta-Guide.com
· Best Node-RED Videos | Meta-Guide.com

30 Aug 2014

= = =

How do I make my own JARVIS?

Check out my quick and dirty webpage:
· 100 Best JARVIS Videos | Meta-Guide.com

See also my recent Quora answer to:
· How do I make a simple robotic servant?

29 Aug 2014

= = =

What is the best natural language processing API / library / service today?

See my recent Quora answer to:
· Which NLP Library is most suitable for use and further development for a text mining startup?

28 Aug 2014

= = =

What is WEKA? How do I get started with it?

Also, check out my quick and dirty webpage:
· 100 Best Weka Tutorial Videos | Meta-Guide.com

28 Aug 2014

= = =

What are some neurobiological models that are useful for NLP applications?

I believe this field is generally known as Biologically inspired cognitive architectures, to which the BICA Society is dedicated.

See their proceedings for details:
· Biologically Inspired Cognitive Architectures 2012
· Biologically Inspired Cognitive Architectures 2011
· Biologically Inspired Cognitive Architectures 2010

28 Aug 2014

= = =

What research is there on artificial intelligence and humor?

=> quora.com/search?q=”computational-humor

According to Wikipedia, the field of study is known as Computational humor. And, there is a Quora topic for Computational Humor. Note, in Commonwealth countries it may be known as “Computational humour”.

For references, see my quick and dirty webpage:
· Computational Humor | Meta-Guide.com

28 Aug 2014

= = =

What is the current relationship between prosody in linguistics, speech recognition, and affective computing?

Quora gives as good an overview of where these things are at as any:

1) quora.com/search?q=prosody

According to Wikipedia, irony is an example of Prosody (linguistics). I suggest looking at the 2010 seminar paper, Verbal Irony: Theories and Automatic Detection. Another example might be the effects of accent on a conversational agent such as Siri, amply illustrated in my own quick and dirty webpage, 100 Best Siri Accent Videos | Meta-Guide.com.

2) quora.com/search?q=speech-recognition

I consider acoustic speech recognition to be primarily a mechanical or hardware solution, and more akin to hard robotics than soft AI, such as NLP. For instance, supposedly the noise cancellation chip in Apple iDevices has allowed them to surpass the speech recognition of most PC hardware, and thus enabled Siri to excel over previous PC based conversational agents.

3) quora.com/search?q=affective-computing

There is a lot of research going into affective computing, and has been a lot of recent discussion about it on Quora. Basically, sentiment analysis is the toe in the water of mainstream affective computing. Generally, sentiment analysis is considered to be an NLP solution.

See also my recent Quora answers to:
· What steps are being taken toward developing artificial emotion in machines, computers, and robots?
· What is the relation between sentiment analysis, natural language processing and machine learning?

28 Aug 2014

= = =

Could Google translate be made to sing?

Theoretically yes, since there is a Google Translate API; however, it would have to be pushed through something like Auto-Tune – which does not seem to have an API yet – via Speech Synthesis or TTS (Text-To-Speech).

27 Aug 2014

= = =

Where can I get into an internship program in NLP in India?

Where are you in India? I have a number of NLP contacts in and around Bengaluru. Message me on Quora, or connect via LinkedIn. BTW, there is a Quora topic for Artificial Intelligence in India.

See also my Quora answer to:
· Will Information Science hold me back? I would still have a chance to do my masters on AI or anything without a “Computer Science” UG degree right?

27 Aug 2014

= = =

What are some resources on the link between AI and strategic decision making?

There doesn’t seem to be a Wikipedia entry for “strategic decision making”. According to AI, this does not compute…. If it’s not in Wikipedia, it does not exist.

For me, “strategic” implies long term, and this is a classic problem with AI, focusing on the long term rather than here and now. Secondly, what are “decisions” and how does AI “make” them? Take IBM Watson for example. How did Watson “decide” the questions for Jeopardy clues? Supposedly,Predictive analytics were the defining advancement of IBM Watson. One thing I can tell you is that “hypothesis generator” is the number one search term on my Meta-Guide.com website.

I suggest starting with the history of Decision support systems and Expert systems.

See also my Quora question:
· What precisely is or is not the relation between the fields of ‘business intelligence’ and ‘artificial intelligence’?

26 Aug 2-14

= = =

What technical events are there related to humanoid robotics?

Ben Goertzel has posted some cool new videos from the 2014 IEEE International Conference on Robotics and Automation (icra2014):
· youtube.com/user/bgoertzel/videos

26 Aug 2014

= = =

What are some good resources for learning about support vector machines (SVM)?

See my quick and dirty webpage:
· 100 Best Support Vector Machine Videos | Meta-Guide.com

26 Aug 2014

= = =

What are the best online free articles to learn AI, data mining and other advanced topics?

IMHO, my own Meta Guide Videography is one of the best sources of easy, quick info on these topics. ;^)

25 Aug 2014

= = =

Which NLP Library is most suitable for use and further development for a text mining startup?

See answers to virtually identical question from 2011:
· Which NLP library among the ones below is most mature and should be used by a startup for its NLP needs?

Edit: > i wanted to ask more specifically which NLP libraries are most development friendly.

In terms of purely development-friendly, it would have to be NLTK in Python; however, more robust production systems seem to be made in Java these days, with OpenNLP on top, and MALLET and Stanford NLP close behind. If it’s got to be Ruby, then look at FreeLing before TreeTagger.

In light of the updated comment, I have presented a metric, see references below:
· NLTK (Natural Language Toolkit) .. Python (240)
· OpenNLP .. Java (38)
· MALLET (MAchine Learning for LanguagE Toolkit) .. Java (34)
· Stanford NLP .. Java (31)
· LingPipe .. Java (5)
· FreeLing .. Ruby (4)
· TreeTagger .. Ruby (2)

References:
· Natural Language Processing Toolkits | Meta-Guide.com

Note: I would like to know which NLP toolkits are missing from this list. And, any feedback on the validity of this Github metric would be useful.

23 Aug 2014

= = =

What is the approximate cost of creating virtual intelligence?

For Virtual Intelligence as defined by Wikipedia, I would guesstimate one million dollars for a functional prototype, and ten million for a scalable deployment.

22 Aug 2014

= = =

What are the big topics in NLP?

Natural language processing is an integral part of the automation of everything, which is predicted to significantly reduce the overall number of jobs within the next generation…. (I keep trying to tell people that robots will make Republicans obsolete.)

One example of this is the natural language generation work of converting data into prose by automatedinsights.com and narrativescience.com. Another example would be any one of the numerous verticals currently being addressed by the new IBM Watson Group cognitive computing.

22 Aug 2014

= = =

I plan to make a Question Answering System for my college using NLP. What could be a good approach towards building this system? Also, how could I structure my database for the same?

Start by reviewing my Quora answers about question answering:
1. What are some good papers to read about Question Answering?
2. How do I make a data-driven question answering system from scratch?
3. Question and answer based search engine using DBpedia?
4. What are the key challenges in designing/developing an effective question answering system?
5. How do I generate questions from corpus or comprehensions using NLP concepts?
6. What percentage of questions would IBM’s Watson be able to answer on Quora?
7. What is the current status of systems which can answer questions framed from a given text article?
8. Are there any web sites that integrate social question-answering with conversational bots?
9. Can Quora Q&A be fed into a robot so that it could answer questions like a human?
10. What are some creative ideas for Quora’s (future) API?
11. How does IBM Watson advance question answering in general?
12. How would you build a robot to answer questions on Quora?
13. What are some examples of generalizable question-answering methods for specialized domains?
14. What’s best way to train an chatbot (e.g. alicebot)?
15. What are some good ways to start on a long AI project?
16. Would an AI be allowed to answer questions on Quora?
17. What are the most popular social Q&A sites in France?
18. Will robots take over Quora?

19 Aug 2014

= = =

Will robots take over Quora?

What are the key challenges in designing/developing an effective question answering system?
Make no mistake, Quora co-founder Adam D’Angelo is widely reported to have invested in Vicarious (company) - to build “software that thinks and learns like a human”….

Would an AI be allowed to answer questions on Quora?
There is a topic for Quora Bots (Quora feature). I messaged Quora founderCharlie Cheever in 2011 suggesting… a voluntary, opt-in “bot-zone” as a sandbox for question answering bot projects. To a large extent working with question answering bots would depend on the as yet non-existentQuora API. However, question answering bots may also conflict with the work of Quora founder Adam D’Angelo on his Vicarious (company)… “building software that thinks and learns like a human”.

What are the most promising companies currently working on software solutions to the Strong AI question?
Even Quora co-founder Adam D’Angelo‘s Vicarious (company) is worth keeping an eye on.

>>Vicarious is developing machine learning software based on the computational principles of the human brain…. Powering this technology is a new computational paradigm we call the Recursive Cortical Network<<

19 Aug 2014

= = =

Are there any startups in India based on instrumentation control and robotics?

AndyRobo.com are doing cool things with modular, cloud robotics in Bengaluru….

18 Aug 2014

= = =

Domotics: How do I make a simple robotic servant?

It’s all in your definitions; but, start by checking out my quick and dirty webpages:
· 100 Best Guile3D Denise Videos | Meta-Guide.com
· 100 Best JARVIS Videos | Meta-Guide.com
· 100 Best Smart Home Voice Control Videos | Meta-Guide.com

Hint, use Guile3D Denise to build your “JARVIS” for smart home voice control….

18 Aug 2014

= = =

Are there any software tools or APIs that will allow you to create a 3D talking head that will generate correct lip movements when given a text stream?

I don’t have a direct answer for you; I don’t know of the perfect solution. I do have some resources which may be of help. Generally in robotics this is referred to as embodiment, even for software robots (softbots). Talking heads may be virtual or physical, as in animatronic or robotic heads. The new virtual mannequins are a bit of a combination of the two, using rear projection.

Haptek.com was a good general purpose, desktop avatar system for experimental AI, but is more or less out of business. Oddcast SitePal seems to be the current industry standard, and offers a number of API packages; see both Oddcast – Developers and SitePal API. Cantoche LivingActor.comwas used to animate the well known viral video, “AI vs. AI. Two chatbots talking to each other”.

Robert Medeksza of Zabaware was developing a new 3D Character Engine at sourceforge.net, circa 2010. The well known German AI lab DFKI offers a realtime character animation engine, EMBR.

See also my previous Quora answer on using the SecondLife virtual world as a testbed for prototyping: I’d like to build a talking animatronic Beethoven bust for my music room, and don’t even know where to begin, how to get started?

Avatar Systems:
· Avatar Systems | Meta-Guide.com
· Best Haptek Videos | Meta-Guide.com
· Best Talking Avatar Videos | Meta-Guide.com
· PIAVCA (Platform Independent API for Virtual Characters and Avatars) | Meta-Guide.com

Talking Heads:
· 100 Best Animatronic Head Videos | Meta-Guide.com
· 100 Best Robot Head Videos | Meta-Guide.com
· Talking Heads (Draft) | Meta-Guide.com

Virtual Mannequins:
· 100 Best Virtual Mannequin Videos | Meta-Guide.com

Facial Animation:
· 100 Best CrazyTalk Tutorial Videos | Meta-Guide.com
· 100 Best Faceposer Videos | Meta-Guide.com
· 100 Best Faceshift Videos | Meta-Guide.com
· Best FaceFX Lipsync Videos | Meta-Guide.com
· Best Mixamo FacePlus Videos | Meta-Guide.com

Lip Synchronization:
· 100 Best Anime Studio Lipsync Videos | Meta-Guide.com
· 100 Best Blender Lipsync Videos | Meta-Guide.com
· 100 Best Flash Lipsync Videos | Meta-Guide.com
· 100 Best Lipsync Tutorial Videos | Meta-Guide.com
· 100 Best ToonBoom Lipsync Videos | Meta-Guide.com
· Best Annosoft Lipsync Videos | Meta-Guide.com
· Best Cinema4D Lipsync Videos | Meta-Guide.com
· Best Daz3D Mimic Lipsync Videos | Meta-Guide.com
· Best Papagayo Lipsync Videos | Meta-Guide.com
· Best SecondLife Lipsync Videos | Meta-Guide.com
· Best Synfig Lipsync Videos | Meta-Guide.com

18 Aug 2014

= = =

Are there any apps like Siri for Android?

For a detailed listing, see my quick and dirty webpage:
· 100 Best Google Play Chatbot Apps | Meta-Guide.com
Between August 2012 and May 2014, chatbots available on Google Play increased from 42x to 71x, which points to more than one new chatbot app appearing per month just for the Android operating system.

Written 17 Aug.

= = =

What is the best tool to summarize a text document?

IMHO, there is no best tool to summarize a text document….

See all my Quora answers on summarization:
· quora.com/search?q=marcus-l-endicott+summarization

Written 14 Aug.

= = =

How do I make a speech recognition system which executes my voice commands? Which language should I use for this?

N.B.

1) Nobody makes all of this from scratch.

2) Programming languages are not as important as algorithms.

For details, see all my Quora answers:
=> quora.com/Marcus-L-Endicott/answers

Written 14 Aug.

= = =

What steps are being taken toward developing artificial emotion in machines, computers, and robots?

Think reverse sentiment analysis…. There are actually quite a lot of strides being made in this field known as affective computing. See the 2011 book, Affective Computing and Sentiment Analysis.

See also my quick and dirty webpages:
· Affective Computing & Dialog Systems 2011 | Meta-Guide.com
· Emotional Agents | Meta-Guide.com
· Emotional Agents 2012 | Meta-Guide.com

Written 14 Aug.

= = =

How can I get high school students interested in natural language processing?

Get them into making socialbots (social media bots), such as:
· Chatbots => chatbots.org/ai_zone
· Twitter bots => botforum.net/forums

Written 14 Aug.

= = =

What is the relation between sentiment analysis, natural language processing and machine learning?

· Sentiment analysis is the most widely implemented, practical application of natural language processing at the moment.
· Natural language processing is application of the tasks of NLP [1][2].
· Machine learning involves a number of techniques, including classification, clustering, and neural networks – primarily to deal with the big datainherent in sentiment analysis and the tasks of NLP.

References:
1. Natural language processing: Major tasks in NLP
2. Category:Tasks of natural language processing

Written 10 Aug.

= = =

Who is the world’s top artificial intelligence system designer?

More like, who are the stand-outs, just off the top of my head….

1) IBM Watson principal investigator: David Ferrucci

2) Siri co-founders: Adam Cheyer, Dag Kittlaus, and Tom Gruber

3) China Brain Project mavericks: Hugo de Garis and Ben Goertzel

4) Blue Brain Project founding director: Henry Markram

Reference:
· People Meta Guide | Meta-Guide.com

Written 10 Aug.

= = =

What could you do with Oculus Rift’s and Google Glass’s love child?

What types of applications would you like to see for the Google AR Glasses?
I would like to see avatars move back and forth between virtual worlds and augmented reality. I want voice-interactive NPC chatbots in augmented reality. I would like to see a *platform* where anyone can launch their own embodied #Siri or #Watson; ideally, this would be some kind of pluggable middleware, where people could mix and match any cloud intelligence, or API, with a myriad of customizable 3D animations ala Second Life.

Written 10 Aug.

= = =

What is the difference between middleware and framework?

A framework is something like a development environment that is primarily characterized by libraries. Frameworks may be used for development in general; whereas, middleware is primarily used for integration purposes. A framework used for integration may be referred to as a middleware framework.

Reference:
· Artificial Intelligence Middleware (Draft) | Meta-Guide.com

Written 10 Aug.

= = =

Robo-Advisor: What is a robo-advisor?

en.wikipedia.org/wiki/Robo-Advisor
Robo-advisors are a class of financial advisor that provides portfolio management online with minimal human intervention. While their recommendations may vary, they are all based on algorithms that originally served the traditional advisory community, which has relied on algorithmic templates to conduct portfolio management since at least 2005.

Written 9 Aug.

= = =

How do I determine if two phrases are equal in meaning in natural language processing, like are “gravitational law” and “law of gravity” semantically equivalent?

=> quora.com/search?q=word2vec

There is a neural network based tool developed at Google called word2vec which apparently uses the vector space model for determining semantic similarity.

See also:
· Word2vec Neural Network | Meta-Guide.com

Written 9 Aug.

= = =

Machine Learning: What does “in the wild” imply in machine learning?

I can’t say specifically for machine learning; however, in general for AI “in the wild” refers to public availability. For instance, with conversational AI that uses machine learning there is a well known dumbing down that happens in the wild, usually from people talking dirty and abusing the bot. An example of this would be Cleverbot, which “learns” from users (recycles is more accurate) — the reason many parents have considered Cleverbot potentially hazardous for children.

Written 6 Aug.

= = =

What are some examples of generalizable question-answering methods for specialized domains?

There are many proprietary examples of hybrid, FAQ-based customer relationship management systems, where the FAQs are derived from “user-contributed content” (a form of collective intelligence). These systems are hybrid or “augmented” in the sense that they combine conversational AI or NLP with a translucent “mechanical turk” model; in other words, the end user does not know when the machine stops and the human input starts, essentially functioning as augmented intelligence for the “turks” on the backend. Such semi-automated systems have been widely discussed in the popular press.

Written 6 Aug.

= = =

What are some project topic ideas for artificial intelligence?

For details, see my Quora answer to:
· What are the current sub problems that I could address in summarization, or what would you like a summarizer to do for you?

Written 6 Aug.

= = =

How do I create an NLP framework/library (like NLTK) from scratch?

IMHO, I doubt this can be done at all by a single person, certainly not well, without at least a PhD in Computational linguistics. The more efficient approach would be to integrate existing natural language APIs, using Python if you prefer. The real question then becomes who will build the ultimate pluggable integration platform or artificial intelligence middleware for non-programmers….

Written 6 Aug.

= = =

How do I create a specific neural network structure using MATLAB’s neural network toolbox?

Check out my quick and dirty webpage:
· Best MATLAB Neural Network Videos | Meta-Guide.com

Written 6 Aug.

= = =

What is the difference between an agent-based system and an artificial intelligence?

The more I dig around in the fields of Multi-agent Systems and Artificial Intelligence, the more I realize how murky and ill-defined things are on the edge of knowledge. It seems even the greatest academics too often talk at cross-purposes, apparently adrift in the sea of jargon….

http://en.wikipedia.org/wiki/Age…
A review of recent literature on individual-based models, agent-based models, and multiagent systems shows that ABMs are used on non-computing related scientific domains including biology, ecology and social science. Agent-based modeling is related to, but distinct from, the concept of multi-agent systems or multi-agent simulation in that the goal of ABM is to search for explanatory insight into the collective behavior of agents obeying simple rules, typically in natural systems, rather than in designing agents or solving specific practical or engineering problems.

When people speak of “an AI” they are usually referring to an as yet hypothetical “strong AI“, increasingly called Artificial General Intelligence.

Written 6 Aug.

= = =

What are some good research topics in natural language processing (NLP)?

· Computational Stylistics

http://en.wikipedia.org/wiki/Sty…
Modern stylometry draws heavily on the aid of computers for statistical analysis, artificial intelligence and access to the growing corpus of texts available via the Internet. Software systems such as Signature (freeware produced by Dr Peter Millican of Oxford University), JGAAP (the Java Graphical Authorship Attribution Program—freeware produced by Dr Patrick Juola of Duquesne University) and Stylene for Dutch (online freeware by Prof Walter Daelemans of University of Antwerp and Dr Véronique Hoste of University of Ghent) make its use increasingly practicable, even for the non-expert.

Written 4 Aug.

= = =

Artificial Intelligence: How close have we reached to make a machine imagine?

See my Quora answer to:
· What algorithms model the imagination in agent or multi-agent systems?

Written 4 Aug.

= = =

I am interested in knowing the differences between novel types of personalized recommendations, like making use of Semantic Web, neural networks, recommendation engines etc. What are the differences?

Check out the 2011 book by Joseph A. Konstan etc:
· Collaborative Filtering Recommender Systems

See also my own quick and dirty webpages:
· 100 Best Recommender System Videos | Meta-Guide.com
· Conversational Recommender Systems | Meta-Guide.com
· Recommender Dialog Systems 2011 | Meta-Guide.com
· Recommender Dialog Systems 2012 | Meta-Guide.com

Written 29 Jul.

= = =

What things do I need to know to build an IBM Watson-like DeepQA system on a small scale?

See my recent Quora answer to:
· How do I build my own Watson?

See all my answers about IBM Watson at:
· quora.com/Marcus-L-Endicott/answers/IBM-Watson

Written 28 Jul.

= = =

What is at the heart of a robot?

Most basically, robot means work….

main-qimg

https://en.wikipedia.org/wiki/R.U.R.

R.U.R. is a 1920 science fiction play in the Czech language by Karel Capek. R.U.R. stands for Rosumovi Univerzální Roboti (Rossum’s Universal Robots). However, the English phrase Rossum’s Universal Robots had been used as the subtitle in the Czech original. It premiered on 25 January 1921 and introduced the word “robot” to the English language and to science fiction as a whole.

Written 26 Jul.

= = =

If all sentences like “how to describe” are extracted from the Internet and their answers/frames are embedded with corresponding video frames/text, isn’t this enough to develop “auto-generating text” from video?

For details, see my Quora answers to:
1. How can one start to develop a system that can learn the subject by itself through large amounts of data from the Internet and create automatic short animated videos to explain things to humans?
2. If an artificial intelligence could take in a well-written screenplay and turn that into a simulated movie, would that be an effective test of human-level AI and natural language processing (NLP)?
3. Is there any website that auto-generates video from a story-based text?

… such a reverse generation system could theoretically be made by crunching all YouTube (say frame by frame) coupled with their audio-to-text transcripts…. But, why would such a system be significant? Conceivably, the human mind converts language to image, and vice versa. This process is likely the origin of metaphor. And, metaphor is likely the basis of dream. So, such a reverse generation system could probably not only be used to generate metaphor, but also to translate or decode not just metaphors but dreams as well.

References:
· SceneMaker | Meta-Guide.com
· Text-to-Image Systems (Draft) | Meta-Guide.com
· TTSCS (Text-to-scene Conversion Systems) | Meta-Guide.com

Written 23 Jul.

= = =

I need the API of airlines for my website. How can I get it?

ProgrammableWeb maintains an extensive listing of Air Travel APIs.

Written 22 Jul.

How are chatbots like Cleverbot programmed?

For details, see my Quora answers to:
· How does Cleverbot work?
· How does the chatbot jabberwacky work?
· What’s the techonology behind Korean chat bot simsimi?

Written 20 Jul.

= = =

Are there any alternatives to IRC which provide the same features but are more user-friendly?

There appear to be an increasing number of commercial IM and/or realtime social networks for corporate use, including but not limited to the following:
· campfirenow.com
· salesforce.com/chatter
· IBM Connections
· jivesoftware.com
· socialcast.com
· yammer.com

Written 16 Jul.

= = =

What are the patterns of different questions?

https://en.wikipedia.org/wiki/Fi…
The Five Ws and H are questions whose answers are considered basic in information-gathering. They are often mentioned in journalism (cf. news style), research, and police investigations.They constitute a formula for getting the complete story on a subject. According to the principle of the Five Ws, a report can only be considered complete if it answers these questions starting with an interrogative word:
· Who is it about?
· What happened?
· Where did it take place?
· When did it take place?
· Why did it happen?
· How did it happen?

Written 16 Jul.

= = =

Will Information Science hold me back? I would still have a chance to do my masters on AI or anything without a “Computer Science” UG degree right?

Can you code? (The only thing I need to see is your Github.) If you are interested in some hands on experience working with an AI project in Bangalore, PM me…. I’m confident practical experience like this will prove helpful on any application for grad school. (And, did I say Github?)

Written 14 Jul.

= = =

Are there any collaborative web scraping and indexing projects?

· commoncrawl.org .. “is a non-profit foundation dedicated to providing an open repository of web crawl data that can be accessed and analyzed by everyone.”

Written 13 Jul.

= = =

How do I generate questions from corpus or comprehensions using NLP concepts?

My answers to the following Quora questions may prove helpful:
· What are some good papers to read about Question Answering?
· Can computers make questions?
· What is the current status of systems which can answer questions framed from a given text article?

See also:
· Al-Yahya, Maha. “Ontology-Based Multiple Choice Question Generation.” The Scientific World Journal 2014 (2014).

Written 13 Jul.

= = =

What are the top hotel APIs/XMLs?

ProgrammableWeb maintains an extensive listing of Hotels APIs.

Written 13 Jul.

= = =

TTS (Text-To-Speech): What is the best text-to-speech open source script for development?

Festival Speech Synthesis System is by far the most popular open source TTS toolkit; however, be aware that “speech synthesis” does not mean speech recognition.

Category:Free speech synthesis software:
· espeak.sourceforge.net
· festvox.org/festival
· freetts.sourceforge.net
· sourceforge.net/projects/gnuspeech

Written 13 Jul.

= = =

What are different NLP tools available to an application developer?

For details, see Wikipedia:
1. Category:Tasks of natural language processing
2. Category:Natural language processing software
3. Category:Natural language processing toolkits

See also:
· List of 30 Natural Language Processing APIs – Mashape Blog

Written 12 Jul.

= = =

Are there any apps like Siri?

See my recent Quora answer to:
· Virtual Assistants (bots): What are some good virtual assistant bots?

Written 10 Jul.

= = =

Artificial Intelligence: How can one start to develop a system that can learn the subject by itself through large amounts of data from the Internet and create automatic short animated videos to explain things to humans?

For details, see my Meta Guide Videography:
· http://meta-guide.com/videography/

See also my Quora answers to:
· Is there any website that auto-generates video from a story-based text?
· If an artificial intelligence could take in a well-written screenplay and turn that into a simulated movie, would that be an effective test of human-level AI and natural language processing (NLP)?

Written 6 Jul.

= = =

I want large number of tweets collected from Twitter. Where can I find help?

In addition to DataSift, you can try sifter.texifter.com for smaller/cheaper quantities of tweets than from DataSift. Gnip and I believe Topsy also sell tweets in quantity.

Written 6 Jul.

= = =

Artificial Intelligence: What is your definition of an AI?

Artificial intelligence is basically willy nilly, stabbing in the dark, trying to model “natural intelligence”.

Written 3 Jul.

= = =

Can we make a robot programmer instead of software engineers to code?

See my recent Quora answer to:
· Is it possible to create software that can create new software by text or speech description?

Written 27 Jun.

= = =

Is it possible to create software that can create new software by text or speech description?

See my quick and dirty webpages:
· NaturalJava | Meta-Guide.com
· NLPA (Natural Language Program Analysis) | Meta-Guide.com
· SLP (Spoken Language Programming) | Meta-Guide.com
· VoiceCode | Meta-Guide.com

Written 27 Jun.

= = =

How do I make a data-driven question answering system from scratch?

See my recent Quora answer to:
· How do I build my own Watson?

In fact, see all my Quora answers:
· Marcus L Endicott’s Answers – Quora

Written 26 Jun.

= = =

How do I build my own Watson?

In particular, see:
· IBM Watson — How to replicate Watson hardware and systems design for your own use in your basement (2011/updated 2014)

See also:
· Lavi Nigam’s answer to How does IBM’s Watson work?

Written 19 Jun.

= = =

Personal Digital Assistants: What ever happened with the digital “personal representatives” that were under development a couple of years back?

See my recent Quora answer:
· What are some good virtual assistant bots?

Note, there are Quora topics for both Virtual Assistants (bots) as well as Virtual Assistants (people). Wikipedia has defaulted to the nomenclature of Intelligent software assistant; although, there are also Wikipedia entries for Automated online assistant and Intelligent personal assistant. Additionally, there is a Category:Intelligent software assistants.

The original conception of Personal digital assistant (aka PDA) would seem to have become subsumed into today’s smartphones and tablets….

As for Australian involvement, The Thinking Head Project was a significant government supported initiative, which produced the Thinking Head Software. There are a number of Australian academics engaged in Computational Humor research, notably Michael Meany (Humour in artificial intelligence agents).

It should also be noted that the Australian chatbot company originally known as MyCyberTwin (later renamed Cognea) has just been bought by IBM Watson:
· IBM Watson Acquires Artificial Intelligence Startup Cognea | TechCrunch

Written 19 Jun.

= = =

Virtual Assistants (bots): What are some good virtual assistant bots?

· Apple Siri
· Artificial Solutions Indigo
· Ask Ziggy
· Cluzee
· Cognitive Code SILVIA
· Creative Virtual V-Person
· Dexetra Iris
· Evi
· Existor Cleverbot
· Expect Labs MindMeld
· Google Now
· IBM Watson
· LG Voice Mate
· Maluuba
· Magnifis Robin
· Microsoft Cortana
· Microsoft Tellme
· Nuance Nina
· Nuance Vlingo
· Pannous Jeannie
· Pannous Voice Actions
· Samsung S Voice
· SimSimi
· Skyvi
· Speaktoit Assistant
· Xprevo
· 24me

Edit: Thanks to Anne Hunt for the addition of Expect Labs MindMeld (API & iPad app).

Updated 11 Jul.

= = =

What are frames in English linguistics?

http://en.wikipedia.org/wiki/Fra…
A frame language is a technology used for knowledge representation in artificial intelligence. Frames are stored as ontologies of sets and subsets of the frame concepts. They are similar to class hierarchies in object-oriented languages although their fundamental design goals are different. Frames are focused on explicit and intuitive representation of knowledge where as objects focus on encapsulation and information hiding. Frames originated in AI research and objects primarily in software engineering. However, in practice the techniques and capabilities of frame and object-oriented languages overlap significantly.

Written 10 Jun.

= = =

Artificial Intelligence: What is the state of the art of ontology of questions?

Step #1: Search Google Scholar

Step #2: Apply Automatic Summarization to Google Scholar results

Step #3: Read quick and dirty webpage at Question Ontology | Meta-Guide.com

Written 4 Jun. Asked to answer by Sofiane Oumaouche.

What are the best resources for learning about natural language generation?

See my various Quora answers regarding Natural Language Generation:
· quora.com/search?q=endicott+natural-language-generation

Written 24 May.

= = =

What is a killer text summarization API that will be able to summarize an article to a headline sized piece of text?

There is no killer text summarization API, yet….

Mashape seems to have the most summarization APIs in one place:
· List of 30+ Summarizer APIs, libraries, and software

See my recent “survey” of Github resources on summarization:
100 Best GitHub: Automatic Summarization | Meta-Guide.com

See also my various Quora answers on summarization:
quora.com/search?q=endicott+summarization

And while you’re about it, why not read the success story of Nick D’Aloisio….

Written 24 May.

= = =

Who are some of the most promising groups pursuing brain emulation (government-backed projects, companies, schools, researchers, individuals, etc.)?

See my Quora answers to:
· Is it theoretically impossible for anyone to gain immortality through living in a machine?
· What is the status and direction of research in artificial brain implants for enhancement of human intelligence?

Written 21 May.

= = =

If an artificial intelligence could take in a well-written screenplay and turn that into a simulated movie, would that be an effective test of human-level AI and natural language processing (NLP)?

See my recent Quora answer on text-to-image systems:
· Is there any website that auto generates video from a novel based text?

No, that would not “be an effective test of human-level AI and natural language processing”….

However, such a reverse generation system could theoretically be made by crunching all YouTube (say frame by frame) coupled with their audio-to-text transcripts…. But, why would such a system be significant? Conceivably, the human mind converts language to image, and vice versa. This process is likely the origin of metaphor. And, metaphor is likely the basis of dream. So, such a reverse generation system could probably not only be used to generate metaphor, but also to translate or decode not just metaphors but dreams as well.

See also:
· quora.com/search?q=endicott+(dream+OR+dreams)
· quora.com/search?q=endicott+(metaphor+OR+metaphors)

Written 18 May.

= = =

Artificial Intelligence: What kind of publications are helpful in developing auto generating romance software?

The level of interest in (number of questions about) Natural language generation appears to be increasing…. There is a Quora topic for Natural Language Generation.

Generally speaking, automated books are created from templates; however, “poetry generation systems” tend to be more random, like scrambled nonsense generated by “Markov systems” (aka “Markov generators”). At one time there was a fad of creating “poetry” by translating something into one language and then translating it back using Google Translate, which changed the words/meaning going each way.

Currently, the two main companies with products in the NLG space areAutomated Insights and Narrative Science, focused on creating narrative from raw data. An interesting factoid is that David Ferrucci, principal investigator of IBM Watson, started his career working on “storytelling machines” (aka “automated storytelling”, “story generation” or “story generator”).

References:
· Grammar Parsing & Natural Language Generation | Meta-Guide.com
· Natural Language Generation | Meta-Guide.com
· Natural Language Generation & Dialog Systems 2010 | Meta-Guide.com
· Natural Language Generation & Dialog Systems 2011 | Meta-Guide.com
· Natural Language Generators & Dialog Systems | Meta-Guide.com
· SPARQL & Natural Language Generation | Meta-Guide.com
· Story Engines | Meta-Guide.com
· Story Generator Algorithms | Meta-Guide.com
· Storytelling Systems | Meta-Guide.com

See also my recent Quora answer about text-to-image systems:
· Is there any website that auto generates video from a novel based text?

Updated 14 May.

= = =

What are some cool tutorials/books for learning about augmented reality?

See my quick and dirty webpage:
· 100 Best Augmented Reality Tutorial Videos | Meta-Guide.com

Written 6 May.

= = =

Will people ever interact with artificial intelligence programs as they would with other people? If so, how will this affect society?

Hello?? Haven’t you seen Her (2013 movie) yet?? ;^)

Written 30 Apr.

= = =

Is there any website that auto-generates video from a story-based text?

In particular, see my recent literature review:
· Text-to-Image Systems (Draft) | Meta-Guide.com

See also my quick and dirty webpages:
· SceneMaker | Meta-Guide.com
· TTSCS (Text-to-scene Conversion Systems) | Meta-Guide.com

Written 28 Apr.

= = =

Where can I find Google Now?

LMGTFY:
1. google.com/landing/now
2. Google Now – Wikipedia
3. 100 Best Google Now Videos | Meta-Guide.com

Written 12 Apr.

= = =

How does one create an interactive artificial intelligence within an app with great design?

1) interactive

“Interactive” means to me the speech layer. The speech layer consists of both speech-in and speech-out. At this point in time, the quality of the speech layer is heavily dependent on the hardware platform.

2) artificial intelligence

There is no definitive “artificial intelligence” at this time, due mainly to the fact we haven’t cracked “natural intelligence” yet. Up to now, winning a Turing test, such as the Loebner Prize, has been heavily dependent on trickery, smoke and mirrors. Human conversation is highly psychological in nature, and therefore often marked by theatricality and so-called mind games.

3) app

“App” in this case means for me animated avatar. Lip sync is a major consideration for avatar animation.

So, “design” and in particular “great design” may mean any number of things. Most pertinent to me is the availability of tools to implement great design. Nuance is by far and away the leader in voice technology; but, what then are the open source alternatives to Nuance tools available to common designers? AI not only consists of data, in fact BIG DATA, but also methods of processing that data such that it can be accessed via natural language; where then are the open source engines for transforming data into knowledge and vice versa? Finally, what are the open source toolkits for designing awesome lipsynced animations across mobile platforms? Tell me this then, and maybe then, we can discuss “great design” for mobile AI….

See also my recent answer to:
· Is there an open source personal assistant for the web, like Speak to It Chrome extension that can be customized?

Written 11 Apr.

= = =

What are the best open source resources for automatic document summarization?

I would like to know the answer to this question myself; but, this is a contest that’s only just begun, much less been definitively decided. Beyond the Wikipedia entry on Automatic summarization and the Quora topic on Automatic Summarization, here are some additional resources:

See summarizers:
· Copernic Summarizer | Meta-Guide.com
· DiaSumm (Dialog Summarizer) | Meta-Guide.com
· GistSumm (GIST SUMMarizer) | Meta-Guide.com
· Open Text Summarizer | Meta-Guide.com

See my quick and dirty webpages:
· Automatic Summarization 2011 | Meta-Guide.com
· Automatic Summarization 2012 | Meta-Guide.com
· Automatic Summarization 2013 | Meta-Guide.com
· Sentence Summarization | Meta-Guide.com
· Summarization | Meta-Guide.com
· Summarization Systems | Meta-Guide.com

See also book Contents:
· Automatic Summarization (2011) .. by Ani Nenkova & Kathleen McKeown
· Proceedings of the Workshop on Automatic Text Summarization 2011 .. Canadian AI, St. John’s, Newfoundland

Written 9 Apr.

= = =

Why hasn’t augmented reality caught on?

Until now hardware has been too expensive and therefore relatively inaccessible. For example, Apple created a whole new software market (for apps) by making the iPhone hardware accessible. Once Google Glass gathers enough competition (in other words knockoffs) and prices come way down, as well as some kind of standards emerge (think Android OS), then we should be able to see some serious uptake and more traction for augmented reality.

See also my answer to:
· What types of applications would you like to see for the Google AR Glasses?

Written 5 Apr.

= = =

Is there an open source personal assistant for the web, like Speak to It Chrome extension that can be customized?

There are different aspects of this to consider. Most natural language “engines” (called “dialog systems”) are text-based. Adding both audio-in and audio-out are additional levels of complexity. And adding an animated avatar system is yet another layer.

The short answer is that I know of no do-it-yourself kit for for doing this on the web. Traditionally such bundled “agents” were primarily for desktop machines (leveraging the Microsoft Speech API, for instance). Generally, both video and audio are web browser dependent, and the Browser wars made it difficult to make things work across all platforms (much like issues making cross-platform mobile apps today). It was only fairly recently that Google’s initially undocumented Chrome-based “Web Speech API” became available to developers.

Pandorabots CallMom™ is one DIY system that leverages the Google Speech API. I believe, CallMom also makes use of Pannous “Voice Actions” (aka Jeannie Voice Actions); both Voice Actions (billed “Siri as a service”) and Jeanie APIs are available via Mashape. See also Pannous on GitHub.

References:
· IMified to be discontinued on September 1, 2012
· Pandorabots CallMom™ Mobile Virtual Assistant
· Verbots Online closing down at the end of August

Written 4 Apr.

= = =

What are the advantages of having virtual assistants over personal assistants in a business?

What I always tell people about e-commerce websites is that you do not need awhiz-bang e-commerce website unless you are already getting more orders than you can deal with otherwise…. The same could be said for robotic “virtual assistants” over human “personal assistants”…. I would say that Virtual Assistants (bots) are more suited to broad and shallow; whereas, Virtual Assistants (people) are better suited for narrow and deep…. ;^) Here broad and shallow means dealing with lots of people with a relatively small knowledgebase, and narrow and deep refers to few people and deep knowledge requirements.

Written 3 Apr. Asked to answer by Anonymous.

= = =

What is the status and direction of research in artificial brain implants for enhancement of human intelligence?

IMHO, Google Glass is as close as we’ve come, in a mainstream sense, to the direction of implants. Certainly everything that could be done with Google Glass can eventually be done with implants of one kind or another.

See book Contents:
· Neuromorphic and Brain-Based Robots (2011) .. by Jeffrey L. Krichmar etc

See also my quick and dirty webpages:
· 100 Best Google Glass Videos | Meta-Guide.com
· Artificial Brains | Meta-Guide.com
· Brain Simulation | Meta-Guide.com
· Robot Brains | Meta-Guide.com

Written 2 Apr.

= = =

Natural Language Processing: What are the best Q&A websites for natural language processing?

Natural language processing” comes in all shapes and sizes. “Sentiment analysis” may be the biggest, practical application of NLP these days. Both “Big data” and the “Internet of Things” will undoubtedly drive more interest in NLP applications of all kinds.

I am most interested in “Dialog system” applications. I have not been able to identify commonly used or readily available dialog system applications based primarily on conventional NLP; however, there do seem to be more and more NLP APIs coming online. For instance, there is now a developerWorks Q&A Forum for the newly available IBM Watson Ecosystem APIs. In terms of traditional “Pattern matching” dialog systems, the “Loebner Prize” crowd holds court on the Robitron Yahoo Group; otherwise, various and sundry chatbot hobbyists (usually computer professionals in real life) may be found on the Chatbots.org “AI Zone” forums.

I would like to learn about more (kinds of) active NLP forums, in particular for practical applications (in other words, other than for pure, academic research).

Written 31 Mar.

= = =

What is “parse selection” in NLP?

“Parse selection is the process of deciding which of a number of possible parses (trees) is the most likely for a given sentence.”
· Support vector machines for parse selection, E Dijkstra – Master’s thesis, Univ. of Edinburgh, 2001

Written 29 Mar.

= = =

Are there any email addresses which are hooked up to a Turing Test via a chat client?

See the recent thread on the Chatbots.org “AI Zone” forum, Email Interfaces to Chatbots.

Written 28 Feb.

= = =

What are the most promising companies currently working on software solutions to the Strong AI question?

I’m not endorsing Peter Voss‘ companies as “most promising”; however, they definitely appear to be in the running, see Launching AGI Inc.

>>Adaptive A.I. Inc & Smart Action Company are spawning a new R&D venture focusing on long-term AGI (Artificial General Intelligence). The new company, AGI Inc., is well funded for several years of significant, focused research and development without being distracted by short-term commercial considerations.<<

Even Quora co-founder Adam D’Angelo‘s Vicarious (company) is worth keeping an eye on.

>>Vicarious is developing machine learning software based on the computational principles of the human brain…. Powering this technology is a new computational paradigm we call the Recursive Cortical Network<<

Like Vicarious, Grok Solutions (formerly Numenta) is working with a “Cortical Learning Algorithm”.

There are many others, mostly in stealth, and the successful ones may even drop off the face of the Earth, never to be heard from again….

Written 8 Feb.

= = =

Is there any NLP library/tool/API for cleansing noisy text (e.g. SMS, Twitter) in English to its correct text?

1) See Wikipedia: Text normalization

2) See the Quora topic: Text Normalization

3) See also Wikipedia: OpenRefine (ex-Google Refine)

Written 4 Feb.