Notes:
Latent semantic analysis (LSA) and latent semantic mapping (LSM) are related techniques used in natural language processing (NLP) to analyze the meaning and relationships of words and documents. However, they are not the same thing and have some key differences:
- LSA is a method for analyzing the relationships between words and documents based on their co-occurrence patterns in a large corpus of text. It is based on the idea that words that occur frequently together in the same context tend to have similar meanings. LSA represents words and documents as vectors in a high-dimensional space, where the dimensions represent the underlying latent (hidden) concepts or themes in the text.
- LSM is a method for visualizing the relationships between words and documents in a two-dimensional space. It is similar to LSA in that it uses co-occurrence patterns to infer the meanings and relationships of words and documents, but it presents the results in a visual format rather than as vectors in a high-dimensional space. LSM typically uses multidimensional scaling (MDS) to map the relationships between words and documents onto a two-dimensional plane, where the distance between words and documents reflects the similarity of their meanings.
Both LSA and LSM are used to analyze the semantics (meaning) of words and documents, but they differ in the way they represent and visualize the results. LSA is a mathematical technique that represents the relationships between words and documents as vectors in a high-dimensional space, while LSM is a graphical technique that represents the relationships in a two-dimensional visual space.
Latent semantic analysis (LSA) and probabilistic latent semantic analysis (PLSA) are both techniques used in natural language processing (NLP) to analyze the meaning and relationships of words and documents. LSA is a method for analyzing the relationships between words and documents based on their co-occurrence patterns in a large corpus of text. It is based on the idea that words that occur frequently together in the same context tend to have similar meanings. LSA represents words and documents as vectors in a high-dimensional space, where the dimensions represent the underlying latent (hidden) concepts or themes in the text.
Probabilistic latent semantic analysis (PLSA) is a probabilistic extension of LSA that allows for the incorporation of additional sources of information, such as word frequencies and document labels, into the analysis. PLSA is based on a probabilistic model of how words and documents are generated, which allows it to estimate the probability that a given word came from a particular document. This can be used to improve the accuracy and interpretability of the LSA results by taking into account additional factors that may influence the co-occurrence patterns of words and documents.
In the context of dialog systems, LSA can be used to analyze the semantics (meaning) of user inputs and system responses in order to improve the understanding and interpretation of the dialog. For example, LSA can be used to identify the underlying concepts or themes in user inputs and responses, and to match those concepts with appropriate system responses. This can be particularly useful in handling ambiguous or vague user inputs, as LSA can help the system disambiguate the meaning of the input by considering the context and co-occurrence patterns of the words used.
LSA can also be used to improve the naturalness and coherence of the dialog by ensuring that the system’s responses are semantically related to the user’s inputs and the previous turns in the dialog. This can help the system generate more appropriate and relevant responses and create a more natural and engaging conversation with the user.
- Conversational informatics is a field of study that focuses on the design and development of conversational systems, such as chatbots, voice assistants, and other types of interactive systems that can engage in natural language conversations with users. This field involves a range of disciplines, including computer science, linguistics, psychology, and social sciences, and aims to understand how people communicate and interact with each other, as well as how to design and build systems that can engage in effective and natural conversations with users.
- Topic generation is the process of identifying and generating topics for discussion or analysis. This can involve a range of activities, such as identifying the main themes or ideas within a body of text or data, extracting key words or phrases, and grouping those words or phrases into coherent topics. Topic generation is often used in natural language processing (NLP) and information retrieval (IR) applications to identify the main themes or concepts within a document or collection of documents, and can be used to improve the accuracy and relevance of search results, recommendation systems, and other applications that rely on understanding the content of text.
Resources:
- deeptutor.org .. advanced intelligent tutoring system for deep understanding of complex science topics
- nlplab.org .. virtual lab, natural language processing laboratory
Wikipedia:
See also:
Hierarchical variational memory network for dialogue generation
H Chen, Z Ren, J Tang, YE Zhao, D Yin – … of the 2018 World Wide Web …, 2018 – dl.acm.org
… ABSTRACT Dialogue systems help various real applications interact with hu- mans in an intelligent natural way. In dialogue systems, the task of dialogue generation aims to generate utterances given previ- ous utterances as contexts …
An Ensemble of Retrieval-Based and Generation-Based Human-Computer Conversation Systems.
Y Song, R Yan, CT Li, JY Nie, M Zhang, D Zhao – 2018 – openreview.net
Page 1. Under review as a conference paper at ICLR 2018 AN ENSEMBLE OF RETRIEVAL-BASED AND GENERATION-BASED HUMAN-COMPUTER CONVERSATION SYSTEMS Anonymous authors Paper under double-blind review ABSTRACT …
Learning joint semantic parsers from disjoint data
H Peng, S Thomson, S Swayamdipta… – arXiv preprint arXiv …, 2018 – arxiv.org
… ural language, and has been useful in question an- swering (Shen and Lapata, 2007), text-to-scene generation (Coyne et al., 2012), dialog systems (Chen et … It does not use latent semantic depen- dency structures, and aims to minimize the sum of training losses from both …
Dialogue act recognition via crf-attentive structured network
Z Chen, R Yang, Z Zhao, D Cai, X He – The 41st International ACM SIGIR …, 2018 – dl.acm.org
… task. Many applications have benefited from the use of au- tomatic dialogue act recognition such as dialogue systems, machine translation, automatic speech recognition, topic identification and talking avatars [20] [14]. One …
Deep Learning in Natural Language Processing
L Deng, Y Liu – 2018 – books.google.com
… Knowledge-based question answering Knowledge graph Limited-memory Broyden–Fletcher– Goldfarb–Shanno Latent semantic indexing Long … gisting evaluation Referenced metric and unreferenced metric blended evaluation routine Spoken dialog system Spoken language …
Towards conversational search and recommendation: System ask, user respond
Y Zhang, X Chen, Q Ai, L Yang, WB Croft – Proceedings of the 27th ACM …, 2018 – dl.acm.org
… KEYWORDS Conversational Search; Conversational Recommendation; Product Search; Dialog Systems; Memory Networks; Personalized Agent … Traditional approaches to dialog systems have mostly been Session 1E: Interactive IR 1 …
Recent trends in deep learning based natural language processing
T Young, D Hazarika, S Poria… – ieee Computational …, 2018 – ieeexplore.ieee.org
… to perform a wide range of natural language related tasks at all levels, ranging from parsing and part- of-speech (POS) tagging, to machine translation and dialogue systems … sentence to create an informative latent semantic representation of the sentence for downstream tasks …
Enhance word representation for out-of-vocabulary on ubuntu dialogue corpus
J Dong, J Huang – arXiv preprint arXiv:1802.02614, 2018 – arxiv.org
… One problem in chat-oriented human- machine dialog system is to reply a message within conversation contexts … The size of the corpus makes it attractive for the exploration of deep neural network modeling in the context of dialogue systems …
Response selection with topic clues for retrieval-based chatbots
Y Wu, Z Li, W Wu, M Zhou – Neurocomputing, 2018 – Elsevier
… Recent researches on conversational agents have two dominated directions, namely task oriented dialog systems and non task oriented chatbots … Conversational agents can be categorized into task-oriented dialogue systems and non-task oriented chatbots …
SKOPE-IT (Shareable Knowledge Objects as Portable Intelligent Tutors): overlaying natural language tutoring on an adaptive learning system for mathematics
BD Nye, PI Pavlik, A Windsor, AM Olney… – … journal of STEM …, 2018 – biomedcentral.com
This study investigated learning outcomes and user perceptions from interactions with a hybrid intelligent tutoring system created by combining the AutoTutor conversational tutoring system with the Assessment and Learning in Knowledge Spaces (ALEKS) adaptive learning system …
An introduction to neural information retrieval
B Mitra, N Craswell – Foundations and Trends® in Information …, 2018 – nowpublishers.com
… Traditional IR models such as Latent Semantic Analysis (LSA) (Deerwester et al., 1990) learn dense vector representations of terms and documents … Deerwester, SC, ST Dumais, TK Landauer, GW Furnas, and RA Harshman. 1990. “Indexing by latent semantic analysis”. JASIS …
On the Vector Representation of Utterances in Dialogue Context
L Pragst, N Rach, W Minker, S Ultes – Proceedings of the Eleventh …, 2018 – aclweb.org
… Rieser, V. and Lemon, O. (2011). Reinforcement learning for adaptive dialogue systems: a data-driven methodol- ogy for dialogue management and natural language gen- eration … A latent semantic model with convolutional- pooling structure for information retrieval …
Data-Driven Language Understanding for Spoken Dialogue Systems
N Mrkši? – 2018 – repository.cam.ac.uk
Page 1. Data-Driven Language Understanding for Spoken Dialogue Systems Nikola Mrkšic Supervisor: Professor Steve Young … Page 19. Chapter 2 Statistical Spoken Dialogue Systems This chapter gives an overview of statistical spoken dialogue systems and their core compo …
Knowledge enhanced hybrid neural network for text matching
Y Wu, W Wu, C Xu, Z Li – Thirty-Second AAAI Conference on Artificial …, 2018 – aaai.org
Page 1. Knowledge Enhanced Hybrid Neural Network for Text Matching Yu Wu, †? Wei Wu, ‡ Can Xu, ‡ Zhoujun Li †?? † State Key Lab of Software Development Environment, Beihang University, Beijing, China ‡ Microsoft …
Rich Short Text Conversation Using Semantic-Key-Controlled Sequence Generation
K Yu, Z Zhao, X Wu, H Lin, X Liu – IEEE/ACM Transactions on Audio …, 2018 – dl.acm.org
… inside each group. To achieve this, we first use truncated SVD [28] to perform latent semantic analysis (LSA) [29] on all sentence embeddings in the same memory block and obtain the projection vectors in the LSA space. The K …
A transfer-learnable natural language interface for databases
W Wang, Y Tian, H Xiong, H Wang, WS Ku – arXiv preprint arXiv …, 2018 – arxiv.org
… Overview of our Approach Our goal is to separate out data specific components and fo- cus on the latent semantic structure in a natural … Slot Filling in Dialogue System Dialogue system aims at communicating with a user in a session with multiple turns of dialogs, where state, or …
Zero-shot adaptive transfer for conversational language understanding
S Lee, R Jha – arXiv preprint arXiv:1808.10059, 2018 – arxiv.org
… For implicit transfer of reusable concepts across domains, we represent slots in a shared latent semantic space by em- bedding the slot description … Deep learning for dialogue systems. In Proceedings of ACL 2017, Tutorial Abstracts, 8–14 …
Information Discovery in E-commerce: Half-day SIGIR 2018 Tutorial
Z Ren, X He, D Yin, M de Rijke – … ACM SIGIR Conference on Research & …, 2018 – dl.acm.org
… [20] W. Lei, X. Jin, Z. Ren, X. He, M.-Y. Kan, and D. Yin. Sequicity: Simplifying task- oriented dialogue systems with single sequence-to-sequence architectures … [28] S. Mukherjee, K. Popat, and G. Weikum. Exploring latent semantic factors to find useful product reviews …
Deep learning for sentiment analysis: A survey
L Zhang, S Wang, B Liu – Wiley Interdisciplinary Reviews: Data …, 2018 – Wiley Online Library
… Due to the nonlinear function h(·) and g(·), the autoencoder is able to learn nonlinear representations, which give it much more expressive power than its linear counterparts, such as principal component analysis (PCA) or latent semantic analysis (LSA) …
Sarcasm Detection on Flickr Using a CNN
D Das, AJ Clark – Proceedings of the 2018 International Conference on …, 2018 – dl.acm.org
… They used a Latent Semantic Analysis (LSA)-based approach to extend the list of indicative hash-tags … 2006. ” Yeah Right”: Sarcasm Recognition for Spoken Dialogue Systems. In Ninth International Conference on Spoken Language Processing …
System for MOOC Users
M Guider, AEE Dialogue – Web and Big Data: APWeb-WAIM …, 2018 – books.google.com
… MOOC Guider: An End-to-End Dialogue System for MOOC Users 273 For a sentence x 1, x2,···, xm with length m, it is first … Hidden states ht of sentence encoder are regarded as memories of different time step for a dialogue, and are used for latent semantic vectors generation …
Smart Entertainment-A Critiquing Based Dialog System for Eliciting User Preferences and Making Recommendations
SG Patil – Natural Language Processing and Information Systems …, 2018 – books.google.com
… Page 463. Smart Entertainment-A Critiquing Based Dialog System 461 on the Movie itself. If … threshold). For our exper- iments, we created 30 clusters of dimension d= 10 using Latent Semantic Analysis using the package Gensim …
MOOC Guider: An End-to-End Dialogue System for MOOC Users
Y Li, Y Zhang – Asia-Pacific Web (APWeb) and Web-Age Information …, 2018 – Springer
… A dialogue system usually consists of several parts including a natural language understanding (NLU) component, a dialogue manager (DM) component … encoder are regarded as memories of different time step for a dialogue, and are used for latent semantic vectors generation …
Smart Entertainment-A Critiquing Based Dialog System for Eliciting User Preferences and Making Recommendations
RR Ramnani, S Sengupta, TR Ravilla… – … on Applications of Natural …, 2018 – Springer
… A typical dialog system [11] consists of 5 main components: 1. Speech Recognition: converts the speech signal to a textual representation. 2 … For our experiments, we created 30 clusters of dimension d = 10 using Latent Semantic Analysis using the package Gensim …
Contextual Topic Modeling For Dialog Systems
C Khatri, R Goel, B Hedayatni, A Metanillou… – arXiv preprint arXiv …, 2018 – arxiv.org
… A major hurdle for open-domain dialog systems is their evaluation [22] as there are many valid responses for any given situation … [16] Thomas Hofmann, “Probabilistic latent semantic anal- ysis,” in Proceedings of the Fifteenth conference on Uncertainty in artificial intelligence …
Intent Detection for code-mix utterances in task oriented dialogue systems
P Jayarao, A Srivastava – arXiv preprint arXiv:1812.02914, 2018 – arxiv.org
… wide range of rule-based approaches also have been successfully implemented to tackle the intent detection issue in dialogue systems [6]. Deep … C. Latent Semantic Analysis (Lsa) LSA takes the tfidf vectorizer a step further, it decomposes it into further separate document-topic …
Latent Semantic Analysis Approach for Document Summarization Based on Word Embeddings
K Al-Sabahi, Z Zuping, Y Kang – arXiv preprint arXiv:1807.02748, 2018 – arxiv.org
Page 1. Latent Semantic Analysis Approach for Document Summarization Based on Word Embeddings … In this paper, we employ word embeddings to improve the weighting schemes for calculating the input matrix of Latent Semantic Analysis method …
Intent Discovery Through Unsupervised Semantic Text Clustering
SB Padmasundari – isca-speech.org
… A dialog system requires to understand the intentions of hu- mans and extract the relevant information and actions from the … ventional topic modeling schemes such as probabilistic latent semantic analysis (pLSA) and latent Dirichlet allocation (LDA) need aggregation of short …
Impact of Auxiliary Loss Functions on Dialogue Generation Using Mutual Information
JS Clair, T Conley, J Kalita – cs.uccs.edu
… Pearson. Landauer, TK; McNamara, DS; Dennis, S.; and Kintsch, W. 2013. Handbook of Latent Semantic Analysis. Psychol- ogy Press … How not to evaluate your dialogue system: An empirical study of unsupervised evaluation met- rics for dialogue response generation …
Analysis of Topic Propagation in Therapy Sessions Using Partially Labeled Latent Dirichlet Allocation
I Chaoua, S Consoli, A Härmä, R Helaoui… – … Workshop on Artificial …, 2018 – Springer
… It follows that the analysis and modeling of these human-to-human dialogues may be useful for the development of AI-based dialogue systems able to … Topic models are a family of probabilistic approaches that aim at discovering latent semantic structures in large documents …
Response selection from unstructured documents for human-computer conversation systems
Z Yan, N Duan, J Bao, P Chen, M Zhou, Z Li – Knowledge-Based Systems, 2018 – Elsevier
… Unlike previous work, LCLR [25] applies rich lexical semantic features that obtained from a wide range of linguistic resources including WordNet, the polarity-inducing latent semantic analysis (PILSA) model and different vector space models …
Form-based Dialogue Structure for Task-oriented Conversations
A Chotimongkol, AI Rudnicky – simulation – cs.cmu.edu
… [8] Raux, A., Langner, B., Black, A. and Eskenazi, M., “LET’S GO: Improving Spoken Dialogue Systems for the Elderly and Non-natives”, Proc … [11] Gorman, JC, Foltz, PW, Kiekel, PA, Martin, MJ, Cooke, NJ, “Evaluation of Latent Semantic Analysis- based measures of team …
Attention mechanism Inspired by human visual attention, the attention mechanism is able to help the neural network learn what to “focus” on when making predictions …
L Deng, Y Liu – Deep Learning in Natural, 2018 – Springer
… Goal-oriented dialog system A goal-oriented dialog system needs to understand a user request and complete a related task with … Latent semantic indexing Latent semantic indexing (LSI) is a dimensionality reduc- tion technique that projects queries and documents into a space …
Improving Dialog Systems Using Knowledge Graph Embeddings
B Carignan – 2018 – curve.carleton.ca
Page 1. Improving Dialog Systems using Knowledge Graph Embeddings by Brian Carignan, BCS … Faculty of Graduate and Postdoctoral Affairs acceptance of the thesis Improving Dialog Systems using Knowledge Graph Embeddings submitted by Brian Carignan, BCS …
Changing the Level of Directness in Dialogue using Dialogue Vector Models and Recurrent Neural Networks
L Pragst, S Ultes – Proceedings of the 19th Annual SIGdial Meeting on …, 2018 – aclweb.org
… prepared to interpret their actual meaning. Further- more, a dialogue system should be able to conform to human expectations by adjust- ing the degree of directness it uses to im- prove the user experience. To reach those goals, we …
Auto-Dialabel: Labeling Dialogue Data with Unsupervised Learning
C Shi, Q Chen, L Sha, S Li, X Sun, H Wang… – Proceedings of the 2018 …, 2018 – aclweb.org
… 2017. Frames: a corpus for adding memory to goal-oriented dialogue systems. In Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue, pages 207– 219 … 2014. A latent semantic model with convolutional-pooling structure for informa- tion retrieval …
A Trustworthy, Responsible and Interpretable System to Handle Chit-Chat in Conversational Bots
P Agrawal, A Suri, T Menon – arXiv preprint arXiv:1811.07600, 2018 – arxiv.org
… Chat detection in an intelligent assistant: Combining task- oriented and non-task-oriented spoken dialogue systems. CoRR abs/1705.00746 … A latent semantic model with convolutional-pooling structure for information retrieval …
Intelligent Tutoring Based on a Context-Aware Dialogue in a Procedural Training Environment
J Paladines, J Ramírez – International Conference on Information Theoretic …, 2018 – Springer
… explanations are compared with a set of expectations (ideal answers) and misconceptions (incorrect answers) by using Latent Semantic Analysis … Beetle II [9] implements an approach based on task-oriented dialogue systems, which is more domain specific than the Autotutor one …
Analysis, discovery and exploitation of open data for the creation of question-answering systems
G Molina Gallego – 2018 – rua.ua.es
… 9 3 Objectives 11 3.1 Designing a functional Dialogue System . . . . . 11 3.2 Providing knowledge to the CA … 26 5.4.2 Latent Semantic Analysis (LSA) . . . . 26 5.4.3 Response generation …
Interpretable Semantic Textual Similarity Using Lexical and Cosine Similarity
G Majumder, P Pakray, DEP Avendaño – … of the Computer Society of India, 2018 – Springer
… This work is related to the field of NLU, which gives an explanatory layer is important, with applications in dialogue system, interactive system and educational system … In: Papers of the AIED-2001 Workshop on Tutorial Dialogue Systems, pp …
Unsupervised Dialogue Act Classification with Optimum-Path Forest
LCF Ribeiro, JP Papa – 2018 31st SIBGRAPI Conference on …, 2018 – ieeexplore.ieee.org
… [24] achieved 78.76% of classification accuracy with a k-Nearest Neighbors classifier using Feature Latent Semantic Analysis and … E. Dialog System Technology Challenges datasets The Dialog System Technology Challenges, previously Di- alog State Tracking Challenge …
Representational Learning in Conversational Agents
S Savi?, M Gnjatovi?, D Miškovi?, B Borovac – researchgate.net
… Intellig. 37 (3), pp. 305–320, 2012. [4] T. Landauer and S. Dumais, “A solution to Plato’s problem: The Latent Semantic Analysis theory of … [16] H. Cuayáhuitl, S. Yu, A. Williamson and J. Carse, Deep Reinforcement Learning for Multi-Domain Dialogue Systems, NIPS Workshop on …
The Diagnosing Behaviour of Intelligent Tutoring Systems
R Bent – 2018 – dspace.library.uu.nl
… To determine whether an answer is complete, Autotutor uses a Latent Semantic Analysis (LSA) score to represent the difference between the student’s answer and the ideal answer. Every question has expectations associated with it …
Deep Learning in Natural
L Deng, Y Liu – Springer
… answering KG Knowledge graph L-BFGS Limited-memory Broyden–Fletcher–Goldfarb–Shanno LSI Latent semantic indexing LSTM … for gisting evaluation RUBER Referenced metric and unreferenced metric blended evaluation routine SDS Spoken dialog system SLU Spoken …
Ontology Based Resource for History Education
D Baeva, D Atanasova – TEM JOURNAL-TECHNOLOGY EDUCATION …, 2018 – ceeol.com
… These systems monitor the lexical matches between user’s queries and modules request – answer written about their knowledge base. Chatbots systems allow realizing dialogue system based on natural language … Technology classification with latent semantic indexing …
Transferring SLU Models in Novel Domains
Y Tang, K Mo, Q Xu, C Zhang, Q Yang – 2018 – openreview.net
… Many state-of-the art dialogue systems follow a learning pipeline that includes components such as spoken language understanding (SLU … There have also been some attempts to learn shared latent semantic representations or model param- eters for multiple domains via …
Unsupervised Post-processing of Word Vectors via Conceptor Negation
T Liu, L Ungar, J Sedoc – arXiv preprint arXiv:1811.11001, 2018 – arxiv.org
… This technique has its root in Latent Semantic Analysis (LSA): Caron (2001) first propose to define the post-processed version of E as ˜EEW := ?:,1:nD p 1:n,1:n, where p is the weighting exponent determining the relative weights assigned to each singular vector of ?:,1:n. While …
A Deep Multiple View Sentence Representation Model for Question Answering
H Li, J Li, W Tian – 2018 37th Chinese Control Conference …, 2018 – ieeexplore.ieee.org
… [3] Shen, Y.; He, X.; Gao, J.; Deng, L.; and Mesnil, G.Latent Semantic Model with Convolutional-Pooling Structure for Information Retrieval … Quantitative and qualitative evaluation of DARPA Communicator spoken dialogue systems …
Open-domain neural conversational agents: The step towards artificial general intelligence
S Arsovski, S Wong, AD Cheok – International Journal of …, 2018 – openaccess.city.ac.uk
… able to pass the Turing Test and move AI research towards full AGI, more research effort towards open- domain dialogue system are presented … the high dimensional space of context variables which in turn alleviating the curse of dimensionality by lowering latent semantic space …
Notes on NAACL 2018
Z Zhu – Machine Learning – ziningzhu.me
… 19 4.1.2 Dialogue system components attempts to solve challenges … The embeddings are calculated using LSA (latent semantic analysis + tf-idf using sklearn), SGNS (skip-gram with negative sampling), GloVe, and PPMI (positive point-wise mutual information) …
Prediction of Negative Symptoms of Schizophrenia from Objective Linguistic, Acoustic and Non-verbal Conversational Cues
D Chakraborty, S Xu, Z Yang, YHV Chua… – … on Cyberworlds (CW …, 2018 – ieeexplore.ieee.org
… analysis of speech impairments re- lated to schizophrenia also employ context-based meth- ods like Latent Semantic Analysis (LSA) and … of Interviews with Schizophrenic Patients,” in Proceedings of the 9th International Workshop on Spoken Dialogue Systems (IWSDS), 2018 …
Associative Conversation Model: Generating Visual Information from Textual Information
Y Ishibashi, H Miyamori – 2018 – openreview.net
… ABSTRACT In this paper, we propose the Associative Conversation Model that generates vi- sual information from textual information and uses it for generating sentences in order to utilize visual information in a dialogue system without image input …
Deep Learning in Conversational Language Understanding
G Tur, A Celikyilmaz, X He, D Hakkani-Tür… – Deep Learning in Natural …, 2018 – Springer
… 2014) or chit-chat (Vinyals and Le 2015) systems approaches: The last hidden layer of the query (in each direction) is supposed to contain a latent semantic representation of the whole input utterance, so that it can be utilized for domain and intent prediction (\(d_k\), \(i_k\)) …
Attention-Based CNN-BLSTM Networks for Joint Intent Detection and Slot Filling
Y Wang, L Tang, T He – … and Natural Language Processing Based on …, 2018 – Springer
… We discuss the advantage of using BLSTM-CNN as an encoder to encode the input sequence for intent detection and slot filling in task-oriented dialog systems in the next experimental analysis … And the local important latent semantic factors can be selected by convolution layer …
Low-Resource Contextual Topic Identification on Speech
C Liu, M Wiesner, S Watanabe, C Harman… – arXiv preprint arXiv …, 2018 – arxiv.org
… One conversation session between user and dialog system, which can be viewed as one spoken document, may include multiple turns, and the user query in each turn is a spoken … Latent Semantic Analysis (LSA) [19] transformation can then be learned from the tf-idf features …
Can You be More Polite and Positive? Infusing Social Language into Task-Oriented Conversational Agents
YC Wang, R Wang, G Tur, H Williams – alborz-geramifard.com
… word2vec is one of the state-of-the-art word embedding methods, which convert each word to a vector representation in a latent semantic space such that words used in common … Building end-to-end dialogue systems using generative hierarchical neural network models …
Investigating the Effects of Word Substitution Errors on Sentence Embeddings
R Voleti, JM Liss, V Berisha – arXiv preprint arXiv:1811.07021, 2018 – arxiv.org
… 2018. [2] Thomas K. Landauer and Susan T. Dumais, “A solution to Plato’s problem: The latent semantic analysis theory … Jung, Cheongjae Lee, Kyungduk Kim, and Gary Ge- unbae Lee, “An integrated dialog simulation technique for evaluating spoken dialog systems,” in Coling …
MOOC-O-Bot: Using Cognitive Technologies to Extend Knowledge Support in MOOCs
A Mitral, L Vigentini, M Djatmiko… – … , and Learning for …, 2018 – ieeexplore.ieee.org
… Comparison is done using latent semantic analysis … In addition, Ferguson [25] showcased a conversational system to solve routing problems in a simple transportation domain and several other such researches present dialogue system for multiple use cases …
Sentiment analysis of peer review texts for scholarly papers
K Wang, X Wan – The 41st International ACM SIGIR Conference on …, 2018 – dl.acm.org
… Memory network is a general machine learning framework introduced by [43] and has made great success in question answer- ing [23, 26, 30, 35, 43], dialogue system [5, 12, 42] and so … [31] used effective data prepro- cessing techniques along with latent semantic analysis and …
Automated essay scoring in applied games: Reducing the teacher bandwidth problem in online training
W Westera, M Dascalu, H Kurvers, S Ruseti… – Computers & …, 2018 – Elsevier
… A breakthrough failed to occur, however, because of the underestimated complexity of creating intelligent dialogue systems … Latent Semantic Analysis (LSA) (Landauer & Dumais, 1997) and Latent Dirichlet Allocation (LDA) (Blei, Ng, & Jordan, 2003) semantic models were …
Explicit retrofitting of distributional word vectors
G Glavaš, I Vuli? – Proceedings of the 56th Annual Meeting of the …, 2018 – aclweb.org
Page 1. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Long Papers), pages 34–45 Melbourne, Australia, July 15 – 20, 2018. c 2018 Association for Computational Linguistics 34 Explicit Retrofitting of Distributional Word Vectors …
The Use of the Convolutional Neural Network as an Emotion Classifier in a Music Recommendation System
PS Lopes, EL Lasmar, RL Rosa… – Proceedings of the XIV …, 2018 – dl.acm.org
… máquina. REFERENCES [1] SN Ahmad. Emotions and topics in online word of mouth: Application of latent semantic analysis. Publicado em CD-ROM, May 2013. [2] LF Brunialti, SM Peres, V. Freire, and CM Lima. Aprendizado …
Hermitian Co-Attention Networks for Text Matching in Asymmetrical Domains.
Y Tay, AT Luu, SC Hui – IJCAI, 2018 – ijcai.org
… In AAAI, 2018. [Lowe et al., 2015] Ryan Lowe, Nissan Pow, Iulian Serban, and Joelle Pineau. The ubuntu dialogue corpus: A large dataset for research in unstructured multi-turn dialogue systems. arXiv preprint arXiv:1506.08909, 2015 …
A Formal Definition of Importance for Summarization
M Peyrard – arXiv preprint arXiv:1801.08991, 2018 – arxiv.org
Page 1. A Formal Definition of Importance for Summarization Maxime Peyrard Research Training Group AIPHES and UKP Lab www.aiphes.tu-darmstadt.de Abstract Research on summarization has mainly been driven by empirical …
Classification of Emoji Categories from Tweet Based on Deep Neural Networks
K Matsumoto, M Yoshida, K Kita – … of the 2nd International Conference on …, 2018 – dl.acm.org
… Therefore, the feature selection method that leave only important words to judge categories of documents as features, or the dimension reduction method such as Latent Semantic Analysis (LSA) [13] and probability Latent Semantic Indexing (pLSI) [14] are used …
Multi-cast attention networks for retrieval-based question answering and response prediction
Y Tay, LA Tuan, SC Hui – arXiv preprint arXiv:1806.00778, 2018 – arxiv.org
Page 1. Multi-Cast Attention Networks for Retrieval-based Question Answering and Response Prediction Yi Tay Nanyang Technological University Singapore ytay017@e.ntu.edu.sg Luu Anh Tuan Institute for Infocomm Research Singapore at.luu@i2r.a-star.edu.sg …
Detecting Unipolar and Bipolar Depressive Disorders from Elicited Speech Responses Using Latent Affective Structure Model
KY Huang, CH Wu, MH Su… – IEEE Transactions on …, 2018 – ieeexplore.ieee.org
… 3.5 Class-Specific Latent Affective Space Model For mood disorder modeling, the emotion profiles ob- tained from the LSTM-based emotion detector are used to construct the class-specific LASM, which adopts latent semantic analysis (LSA), to model the structural relation …
Content-Based Table Retrieval for Web Queries
Y Sun, Z Yan, D Tang, N Duan, B Qin – Neurocomputing, 2018 – Elsevier
Skip to main content …
Deep code comment generation
X Hu, G Li, X Xia, D Lo, Z Jin – Proceedings of the 26th Conference on …, 2018 – dl.acm.org
… IR approaches such as Vector Space Model (VSM) and Latent Semantic Indexing (LSI) usually search comments from similar code snippets … Seq2Seq model is widely used for machine translation [37], text summarization [34], dialogue system [39], etc …
Robots on stage: A cognitive framework for socially interacting robots
I Rodriguez, A Astigarraga, E Lazkano… – Biologically inspired …, 2018 – Elsevier
… Fig. 3 shows the architecture of our dialogue system that incorporates an Automatic Speech Recognizer (ASR), Language Interpreter, Dialogue Manager, Response Selector, Text Generator, Speech Synthesizer (TTS) and the Singing Synthesizer (TTSKantari) …
Resolving Abstract Anaphora Implicitly in Conversational Assistants using a Hierarchically stacked RNN
P Khurana, P Agarwal, G Shroff, L Vig – Proceedings of the 24th ACM …, 2018 – dl.acm.org
… ABSTRACT Recent proliferation of conversational systems has resulted in an increased demand for more natural dialogue systems, capable of more … Effectively, the maxpool- ing operation acts as an attention mechanism on the most important latent semantic factors …
Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing
E Riloff, D Chiang, H Julia, T Jun’ichi – Proceedings of the 2018 …, 2018 – aclweb.org
… She works on spoken language processing and NLP, studying text-to-speech synthesis, spoken dialogue systems, entrainment in conversation, detection of deceptive and emotional speech, hedging behavior, and linguistic code-switching (language mixing). xxii Page 23 …
Neural Ideal Point Estimation Network
K Song, W Lee, IC Moon – Thirty-Second AAAI Conference on Artificial …, 2018 – aaai.org
… 2009), and Yupeng et al. (2014) proposed a topic-factorized ideal point model (TFIPM) (Gu et al. 2014) with probabilis- tic latent semantic analysis (PLSA) (Hofmann 1999) to esti- mate the ideal points of legislators based on roll-call data …
Multi-Cast Attention Networks
Y Tay, LA Tuan, SC Hui – Proceedings of the 24th ACM SIGKDD …, 2018 – dl.acm.org
Page 1. Multi-Cast Attention Networks Yi Tay Nanyang Technological University Singapore ytay017@e.ntu.edu.sg Luu Anh Tuan Institute for Infocomm Research Singapore at.luu@i2r.a-star.edu.sg Siu Cheung Hui Nanyang …
Chinese metaphor sentiment analysis based on attention-based LSTM
Y Peng, C Su, Y Chen – 2018 Tenth International Conference …, 2018 – ieeexplore.ieee.org
… Metaphor computing is helpful in many natural language processing tasks, such as opinion mining, discourse understanding, and dialogue system … 50–55. [34] C. Su, J. Tian, and Y. Chen, “Latent semantic similarity based inter- pretation of chinese metaphors,” Engineering …
COTA: Improving the Speed and Accuracy of Customer Support through Ranking and Deep Networks
P Molino, H Zheng, YC Wang – Proceedings of the 24th ACM SIGKDD …, 2018 – dl.acm.org
… Similarly, existing research in spoken dialogue systems aims to build models to detect intent and extract named entities for call classification and routing [eg, 10, 25, 29] … We refer to this task as contact type identification (similar to intent de- tection in dialogue systems research) …
Efficient Extraction of Named Entities from New Domains Using Big Data Analytics
CJ Saju, S Ravimaran – Journal of Computational and …, 2018 – ingentaconnect.com
… Modeling,29 website translations,33 Sentiment analysis,36 Digital Assistants,31 Speech Recognition,27 Dialog systems,34 Spelling … learning algorithms such as CRF (Conditional Random Fields), LDA (Latent Dirich- let Allocation), LSI (Latent Semantic Indexing), HMM (Hidden …
The Design and Implementation of XiaoIce, an Empathetic Social Chatbot
L Zhou, J Gao, D Li, HY Shum – arXiv preprint arXiv:1812.08989, 2018 – arxiv.org
… The development of social chatbots, or intelligent dialogue systems that are able to engage in empathetic conversations with humans, has been one of the longest running goals in Artificial Intelligence (AI) … Dialogue Manager is the central controller of the dialogue system …
MIX: Multi-Channel Information Crossing for Text Matching
H Chen, FX Han, D Niu, D Liu, K Lai, C Wu… – Proceedings of the 24th …, 2018 – dl.acm.org
… Yu Xu Mobile Internet Group, Tencent henrysxu@tencent.com ABSTRACT Short Text Matching plays an important role in many natural lan- guage processing tasks such as information retrieval, question an- swering, and dialogue systems …
On Character vs Word Embeddings as Input for English Sentence Classification
J Hammerton, M Vintró, S Kapetanakis… – Proceedings of SAI …, 2018 – Springer
… become an important application of text classification for tasks such as sentiment analysis, as well as question answering and dialogue systems … text-classification.pdf. 4. Deerwester, S., Dumais, ST, Furnas, GW, Landauer, TK, Harshman, R.: Indexing by latent semantic analysis …
Personalizing Search Results Using Hierarchical RNN with Query-aware Attention
S Ge, Z Dou, Z Jiang, JY Nie, JR Wen – Proceedings of the 27th ACM …, 2018 – dl.acm.org
… [27] employed HRNN to deal with cross- and in- session commodity information in a session-based recommendation. Besides, hierarchical recurrent encoder-decoder was designed to handle text generation, such as building dialogue system [28] and query suggestions [34] …
KIT-Conferences
MIAR Roedder – 2018 – isl.anthropomatik.kit.edu
… 2nd Workshop on Neural Machine Translation and Generation – WNMT 2018. An End-to-End Goal-Oriented Dialog System with a Generative Natural Language Response Generation … International Workshop on Spoken Dialogue Systems Technology – IWSDS 2018 …
A Bi-Encoder LSTM Model for Learning Unstructured Dialogs
D Shekhar – 2018 – digitalcommons.du.edu
… tems or Conversational Agents – perhaps a desirable application of the future- have been growing rapidly. A Dialog System can communicate with human in text, speech or both and can be classified into – Task-oriented Systems and Chatbot Systems …
Interpretable multimodal retrieval for fashion products
L Liao, X He, B Zhao, CW Ngo, TS Chua – 2018 ACM Multimedia …, 2018 – dl.acm.org
… 2 RELATED WORK 2.1 Fashion retrieval Interest in fashion retrieval has increased recently. While text re- trieval looks for repetitions of query words in text descriptions or product titles, newer latent semantic models [2, 34] use more pow- erful distributed representations [11] …
Web forum retrieval and text analytics: A survey
D Hoogeveen, L Wang, T Baldwin… – … and Trends® in …, 2018 – nowpublishers.com
Page 1. Preprint Foundations and Trends R in Information Retrieval Vol. 12, No. 1 (2018) 1–163 c 2018 D. Hoogeveen, L. Wang, T. Baldwin, KM Verspoor DOI: 10.1561/1500000062 Web Forum Retrieval and Text Analytics: a Survey …
Neural Argument Generation Augmented with Externally Retrieved Evidence
X Hua, L Wang – arXiv preprint arXiv:1805.10254, 2018 – arxiv.org
… Therefore, we design a novel eval- uation method to measure whether the generated arguments contain topic-relevant information. To achieve the goal, we first train a topic- relevance estimation model inspired by the latent semantic model in Huang et al. (2013) …
A Virtual Chatbot for ITSM Application
S Raut – … FOR CONVERGENCE IN TECHNOLOGY (AJCT)-UGC …, 2018 – asianssr.org
… proposed a solution for failing cases of Artificial Intelligence Markup Language (AIML) by adding a Latent Semantic analysis (LSA … Iulian V. Serban, Alessandro Sordoni, Yoshua Bengio, Aaron Courville, Joelle Pineau, “Building End-To-End Dialogue Systems Using Generative …
A Comparison of Features for the Automatic Labeling of Student Answers to Open-ended Questions
JG Alvarado, HA Ghavidel, A Zouaq, J Jovanovic… – pdfs.semanticscholar.org
… Among multiple elements in our data set, our experiments are based only on the labeled student responses to the survey and model answers (expected answers to the questions). Student SAQ responses and associated metadata were collected through a dialog system …
A Comparison of Features for the Automatic Labeling of Student Answers to Open-Ended Questions.
JGA Mantecon, HA Ghavidel, A Zouaq… – … Educational Data Mining …, 2018 – ERIC
… Among multiple elements in our data set, our experiments are based only on the labeled student responses to the survey and model answers (expected answers to the questions). Student SAQ responses and associated metadata were collected through a dialog system …
Language, emotion, and the emotions: A computational introduction
D Santos, B Maia – Language and Linguistics Compass, 2018 – Wiley Online Library
… Bestgen and Vincze (2012) computed the correlation between the emotions associated to the words alone (in ANEW) and the emotions referred to in text (obtained by applying latent semantic analysis and assigning the average of the 30 ANEW?rated closest words in a 10 …
Test of English Language Learning (TELL)(Pearson, 2016)
N Gokturk – Language Assessment Quarterly, 2018 – Taylor & Francis
… For extended constructed oral/written responses, several scoring models, such as Latent Semantic Analysis, an algorithm developed to measure … Spoken Dialog Systems, in which computers act as conversational agents, could possibly be crafted for creating interactive tasks …
Semantic decomposition and marker passing in an artificial representation of meaning
DG Design – Computer Communications, 2018 – masp.dai-labor.de
Social Media Analysis based on Semanticity of Streaming and Batch Data
BG HB – arXiv preprint arXiv:1801.01102, 2018 – arxiv.org
… Distributional semantics paves way to advances in research in cognitive science by in- cluding statistical features of word distribution along with traditional semantic features utilized in Latent Semantic Analysis [20][21]. It is clear that sexual aspects and vocab …
Intelligent Tutoring Systems: A Comprehensive Historical Survey with Recent Developments
A Alkhatlan, J Kalita – arXiv preprint arXiv:1812.09628, 2018 – arxiv.org
Page 1. Intelligent Tutoring Systems: A Comprehensive Historical Survey with Recent Developments ALI ALKHATLAN, University of Colorado Colorado Springs, USA JUGAL K. KALITA, University of Colorado Colorado Springs …
Neural Random Projections for Language Modelling
D Nunes, L Antunes – arXiv preprint arXiv:1807.00930, 2018 – arxiv.org
… matrix factorisation. In (Papadimitriou et al., 1998), ran- dom projections are used as a first step for Latent Semantic Analysis (LSA), which is essentially a matrix factorisation of word co-occurrence counts (Landauer & Dumais, 1997) …
ASHuR: Evaluation of the Relation Summary-Content Without Human Reference Using ROUGE
A Ramírez-Noriega, R Juárez-Ramírez, S Jiménez… – Computing and …, 2018 – cai.sk
… In [11] the authors proposed an integrated method to evaluate summaries using Latent Semantic Analysis (LSA) automatically. This method is based on a regression equation calculated with a corpus of a hundred summaries. It is validated on a dif- ferent sample of summaries …
Response Generation For An Open-Ended Conversational Agent
N Dziri – 2018 – era.library.ualberta.ca
… 26 2.4.2 Task-oriented dialogue systems … 36 LDA Latent Dirichlet Allocation. 6, 24, 25, 50, 53, 56, 58, 73 LM Language Models. 16, 18 LSA Latent Semantic Analysis. 24 LSTM Long Short Term Memory. 19–21 MDP Markov Decision Process. 34 …
Abstractive and Extractive Text Summarization using Document Context Vector and Recurrent Neural Networks
C Khatri, G Singh, N Parikh – arXiv preprint arXiv:1807.08000, 2018 – arxiv.org
… mapping to headlines/phrase representation) [1, 7]. Even though Seq2Seq mod- els are providing benchmark results in Machine Translation and Speech Recognition tasks [6, 35, 38] they have not yet performed well for summarization tasks, dialog systems and evaluation of di …
Group Cognition and Collaborative AI
J Koch, A Oulasvirta – Human and Machine Learning, 2018 – Springer
… Examples include interactive health interfaces [33, 71] and industrial workflows [39], along with dialogue systems such as chat bots [41] and … After every answer, it compares the response with the objectives by applying latent semantic analysis, then chooses its communication …
MathBot: Transforming Online Resources for Learning Math into Conversational Interactions
J Grossman, Z Lin, H Sheng, JTZ Wei, JJ Williams… – footprints.stanford.edu
Page 1. MathBot: Transforming Online Resources for Learning Math into Conversational Interactions Joshua Grossman? Stanford University jgrossman@ stanford.edu Zhiyuan Lin? Stanford University zylin@cs.stanford.edu …
Interactive Spoken Content Retrieval by Deep Reinforcement Learning
HY Lee, PH Chung, YC Wu, TH Lin… – IEEE/ACM Transactions on …, 2018 – dl.acm.org
… In the past development of dia- logue systems, much experience has been accumulated about human-machine interaction, resulting in numerous successful spoken dialogue systems in the past decades … Typical dialogue system defines a set of hand-crafted states …
To Read or To Do? That’s The Task
Z Alibadi, J Vidal – jmvidal.cse.sc.edu
… processing (NLP) enables computers to perform a wide range of natural language-related tasks such as parsing, part-of-speech (POS) tagging, machine translation, dialog systems, and sentiments … grams to create a useful latent semantic representation of the sentence [27] …
Learning proactive behavior for interactive social robots
P Liu, DF Glas, T Kanda, H Ishiguro – Autonomous Robots, 2018 – Springer
… action using a timing threshold. This assumption has been made in HRI (Thomaz and Chao 2011; Chao and Thomaz 2011) and other spoken dialogue systems as well (Raux and Eske- nazi 2008). To determine a time threshold …
Simple Convolutional Neural Networks with Linguistically-Annotated Input for Answer Selection in Question Answering
R Sequiera – 2018 – uwspace.uwaterloo.ca
… That is, they use latent semantic models to capture the hidden word-alignment structures … a new family of Recurrent Neural Networks called a Context- dependent Additive Recurrent Neural Network (CARNN), which is aimed at developing better dialogue systems by considering …
Staqc: A systematically mined question-code dataset from stack overflow
Z Yao, DS Weld, WP Chen, H Sun – arXiv preprint arXiv:1803.09371, 2018 – arxiv.org
Page 1. StaQC: A Systematically Mined Question-Code Dataset from Stack Overflow Ziyu Yao † , Daniel S. Weld # , Wei-Peng Chen †† , Huan Sun † †The Ohio State University, #University of Washington, ††Fujitsu Labs of America …
Catering to Your Concerns: Automatic Generation of Personalised Security-Centric Descriptions for Android Apps
T Wu, L Tang, Z Xu, S Wen, C Paris, S Nepal… – arXiv preprint arXiv …, 2018 – arxiv.org
Page 1. arXiv:1805.07070v1 [cs.CR] 18 May 2018 Catering to Your Concerns: Automatic Generation of Personalised Security-Centric Descriptions for Android Apps Tingmin Wu Swinburne University of Technology tingminwu@swin.edu.au …
A Cognitive Assistant for improving human reasoning skills
NT Le, L Wartschinski – International Journal of Human-Computer Studies, 2018 – Elsevier
… LIZA 2 is designed according to the typical architecture of dialog systems (Lester, Branting, Mott, 2004, Masche, Le, 2017), which usually consist of the following components: a dialog manager ControlUnit for controlling the flow of a conversation between the dialog system and …
Bidirectional Attentional Encoder-Decoder Model and Bidirectional Beam Search for Abstractive Summarization
K Al-Sabahi, Z Zuping, Y Kang – arXiv preprint arXiv:1809.06662, 2018 – arxiv.org
Page 1. 1 Bidirectional Attentional Encoder-Decoder Model and Bidirectional Beam Search for Abstractive Summarization Kamal Al-Sabahi1, Zhang Zuping1*, Yang Kang1 1.School of Information Science and Engineering, Central …
Matrix Factorization Methods For Training Embeddings In Selected Machine Learning Problems
A Fonarev, I Oseledets – skoltech.ru
… 55 3.2.3 Text Embeddings via Probabilistic Latent Semantic Analysis . . 55 9 Page 10 … rithms are widely used in applications related to text meaning understanding, eg, in automated dialog systems. There are a lot of paper that use embeddings for the coref- 19 Page 20 …
Multi-document summarization based on document clustering and neural sentence fusion
TA Fuad – 2018 – opus.uleth.ca
… mating continuous representations of words Latent Dirichlet Allocation (LDA) (Blei et al., 2003) and Latent Semantic Analysis (LSA) (Landauer et al., 1998) are two such examples. The term “Word Embedding” was first introduced in Bengio et al. (2003) where a word em …
Classical and modern Arabic corpora
E Atwell – Diachronic Corpora, Genre, and Language Change, 2018 – books.google.com
… Page 89. ?? Eric Atwell Abu Shawar, Bayan & Atwell, Eric. 2016. Usefulness, localizability, humanness, and language- benefit: Additional evaluation criteria for natural language dialogue systems. International Journal of Speech Technology 19 (2): 373–383. https://doi …
Scalable and Efficient Probabilistic Topic Model Inference for Textual Data
M Magnusson – 2018 – books.google.com
… Doing research at Cornell for one semester really helped me to get different perspectives on the latent semantic analysis research field … 21 3 Probabilistic latent semantic modeling of text 27 3.1 Modelingsemantics …
Low Resource Efficient Speech Retrieval
C Liu – 2018 – jscholarship.library.jhu.edu
… KWS Keyword Search LDA Linear Discriminant Analysis LDC Linguistic Data Consortium LORELEI Low Resource Languages for Emergent Incidents LSA Latent Semantic Analysis LSTM Long Short-Term Memory LVCSR Large Vocabulary Continuous Speech Recognition …
State-of-the-Art Approaches for German Language Chat-Bot Development
N Boisgard – 2018 – ec.tuwien.ac.at
… conversational systems. 2.2.1 Conversational Systems Conversational system, also known as dialog systems, are computer programs which communicate with users using natural language [Jurafsky and Martin, 2017a]. They fall …
Online comments of multi-category commodities based on emotional tendency analysis
X Zhao, C Huang, H Pan – Cluster Computing – Springer
As the promotion competition of comprehensive e-commerce platforms becomes increasingly keener, this paper aims at finding the differences and key elements of consumer perceptions of different…
Discriminative and Adaptive Training for Robust Speech Recognition and Understanding
Z Meng – 2018 – smartech.gatech.edu
… 55 4.4 Latent Semantic Rational Kernel for Topic Spotting … semantic error cost of all possible word sequences on the lattices is minimized given the reference. The semantic error cost between a pair of words can be estimated via latent semantic analysis (LSA) …
Smart and Innovative Trends in Next Generation Computing Technologies: Third International Conference, NGCT 2017, Dehradun, India, October 30-31 …
P Bhattacharyya, HG Sastry, V Marriboyina, R Sharma – 2018 – books.google.com
Page 1. Pushpak Bhattacharyya Hanumat G. Sastry Venkatadri Marriboyina Rashmi Sharma (Eds.) Communications in Computer and Information Science 827 Smart and Innovative Trends in Next Generation Computing Technologies …
Recurrence Quantification Models of Human Conversational Grounding Processes: Informing Natural Language Human-Computer Interaction
CD Rothwell – 2018 – rave.ohiolink.edu
… Conversational grounding is also a crucial process for human-computer interaction using language-based methods, such as spoken dialogue systems … 2013, 2004; Clark and Krych, 2004), making it a high priority for increasing capabilities of spoken dialogue systems …
Inspecting and Directing Neural Language Models
T Noraset – 2018 – search.proquest.com
Page 1. NORTHWESTERN UNIVERSITY Inspecting and Directing Neural Language Models A DISSERTATION SUBMITTED TO THE GRADUATE SCHOOL IN PARTIAL FULFILLMENT OF THE REQUIREMENTS for the degree DOCTOR OF PHILOSOPHY …
Tackling Sequence to Sequence Mapping Problems with Neural Networks
L Yu – arXiv preprint arXiv:1810.10802, 2018 – arxiv.org
Page 1. Tackling Sequence to Sequence Mapping Problems with Neural Networks Lei Yu Mansfield College University of Oxford A thesis submitted for the degree of Doctor of Philosophy Trinity 2017 arXiv:1810.10802v1 [cs.CL] 25 Oct 2018 Page 2 …
Learning Representations of Text through Language and Discourse Modeling: From Characters to Sentences
Y Jernite – 2018 – search.proquest.com
… representing two words can give some information on how they relate to each other se- mantically and morphologically [Mikolov et al., 2013] (similar to ones obtained with document-level non-neural techniques such as Latent Semantic Analysis [Deerwester et al., 1990]) …
Learning to Interpret and Apply Multimodal Descriptions
T Han – 2018 – pub.uni-bielefeld.de
… Moreover, I show that abstract deictic gestures not only lead to better understanding of spatial descriptions, but also result in earlier correct decisions of the system, which can be used to trigger immediate reactions in dialogue systems …
Linguistic alignment classification and generation with deep learning in Spanish conversation
??? – 2018 – s-space.snu.ac.kr
… NLP research enables computers to carry out various tasks such as machine translation and dialogue systems. In particular, when a … main, and can be used to create a dialogue system that generates aligned responses based on user’s speech …
Detecting New, Informative Propositions in Social Media
N Dewdney – 2018 – etheses.whiterose.ac.uk
Page 1. DOCTORAL THESIS Detecting New, Informative Propositions in Social Media Author: Nigel Dewdney Supervisor: Prof. Robert GAIZAUSKAS A thesis submitted in fulfillment of the requirements for the degree of Doctor of Philosophy …
Detecting Recovery Problems Just in Time: Application of Automated Linguistic Analysis and Supervised Machine Learning to an Online Substance Abuse …
R Kornfield, PK Sarma, DV Shah… – Journal of medical …, 2018 – ncbi.nlm.nih.gov
People recommendation on social media
I Guy – Social Information Access, 2018 – Springer
… Latent semantic analysis (LSA) was not shown to produce better results and was not applied since it does not yield intuitive explanations; (2) Content plus Link (CplusL) – combined content-based and graph-based techniques …
Mixture Models for Personalized Recommendation and Prediction
D Kotzias – 2018 – escholarship.org
Page 1. UC Irvine UC Irvine Electronic Theses and Dissertations Title Mixture Models for Personalized Recommendation and Prediction Permalink https://escholarship.org/uc/item/ 48z326rw Author Kotzias, Dimitrios Publication Date 2018 Peer reviewed|Thesis/dissertation …
Short Answer Assessment in Context: The Role of Information Structure
R Ziai – d-nb.info
Page 1. Short Answer Assessment in Context: The Role of Information Structure Dissertation zur Erlangung des akademischen Grades Doktor der Philosophie in der Philosophischen Fakultät der Eberhard Karls Universität T¨ubingen vorgelegt von Ramon Ziai aus Straßburg …
Source Separation and Machine Learning
JT Chien – 2018 – books.google.com
Page 1. Source Separation andMachine Learning JEN-TZUNG CHIEN Page 2. Source Separation and Machine Learning Page 3. This page intentionally left blank Page 4. Source Separation and Machine Learning Jen-Tzung Chien National Chiao Tung University Page 5 …
Representation learning for natural language
O Mogren – 2018 – mogren.one
… humans (Turing 1950). This was published long before machines were anywhere near being able to succeed at this, while substantial progress has been made in recent years using dialog systems trained on large corpora. Most of …