Text Classification & Chatbots

Notes:

Automatic text classification is the process of using machine learning algorithms to automatically assign labels or categories to text data based on its content. This is often done by using a pre-labeled training set of data, where each piece of text has been manually assigned a label or category, and then using that training set to train a machine learning model to make predictions about the label or category of new, unlabeled text data. There are several different machine learning algorithms that can be used for automatic text classification, including naive Bayes, support vector machines, artificial neural networks, and hybrid approaches that combine multiple algorithms.

Automatic text classification plays a vital role in a number of applications, including text summarization, question answering, and information extraction. It can be used to categorize text data for easier organization and analysis, and can also be used to extract specific information or insights from large volumes of text data. For example, automatic text classification could be used to classify customer reviews of a product as positive, negative, or neutral, or to extract specific information from news articles, such as the names of people or organizations mentioned in the article.

Text classification is commonly used in chatbots to understand and interpret the meaning of user input. When a user interacts with a chatbot, they may ask a question or make a statement that the chatbot needs to understand in order to provide an appropriate response. To do this, the chatbot may use text classification techniques to analyze the user’s input and assign it to a predefined category or label. For example, if a user asks a chatbot about the weather, the chatbot might use text classification to determine that the user is asking for information about the weather, and then use a weather API or other data source to provide an answer.

Text classification can be used in chatbots to perform a wide range of tasks, including:

  • Determining the intent of the user’s message: Chatbots can use text classification to understand the overall purpose or goal of the user’s message, such as whether the user is asking for information, making a request, or expressing a desire.
  • Extracting specific information: Chatbots can use text classification to identify and extract specific pieces of information from a user’s message, such as names, dates, or locations.
  • Providing personalized responses: Chatbots can use text classification to tailor their responses to a user’s specific needs or interests, based on the content of the user’s message.
  • Detecting and addressing inappropriate or offensive language: Chatbots can use text classification to identify and respond to inappropriate or offensive language used by users, either by providing a warning or by blocking the message.

Automatic text classification is the process of using machine learning algorithms to automatically assign labels or categories to text data based on its content. This is often done by using a pre-labeled training set of data, where each piece of text has been manually assigned a label or category, and then using that training set to train a machine learning model to make predictions about the label or category of new, unlabeled text data.

Term selection is the process of identifying and selecting specific terms or words from a body of text that are relevant to a particular task or application. This can involve selecting terms that are frequently used, that are characteristic of a particular topic or concept, or that are otherwise deemed important or useful for a given purpose.

Text categorization is the process of assigning text data to predefined categories or labels based on its content. This can be done manually, by a person reading and labeling the text, or automatically, using machine learning algorithms to analyze the text and make predictions about its category.

Text classifier is a machine learning model that is trained to classify text data into predefined categories or labels based on its content. Text classifiers are often used for automatic text classification tasks, and can be trained using a variety of machine learning algorithms, such as naive Bayes, support vector machines, or artificial neural networks.

Resources:

Wikipedia:

References:

See also:

100 Best GitHub: Text Segmentation | NPCEditor | Stanford Classifier | Statistical Classification & Dialog Systems

Developing Enterprise Chatbots
B Galitsky – 2019 – Springer
… 47 3 Explainable Machine Learning for Chatbots ….. 53 3.1 What Kind of Machine Learning a Chatbot Needs … 119 5 Assuring Chatbot Relevance at Syntactic Level . . . … 144 5.5 Evaluation of Text Classification Problems …

An evaluation dataset for intent classification and out-of-scope prediction
S Larson, A Mahendran, JJ Peper, C Clarke… – arXiv preprint arXiv …, 2019 – arxiv.org
… Out-of-scope queries are inevitable for a task- oriented dialog system, as most users will not be fully cognizant of the … et al., 2013), DSTC 2 (Hen- derson et al., 2014a), and DSTC 3 (Henderson et al., 2014b) contain “chatbot style” queries … Bag of tricks for efficient text classification …

When to Talk: Chatbot Controls the Timing of Talking during Multi-turn Open-domain Dialogue Generation
T Lan, X Mao, H Huang, W Wei – arXiv preprint arXiv:1912.09879, 2019 – arxiv.org
… Table 1: In this case, chatbot and human play the role B. The existing chatbots which … et al., 2017): (1) Open-domain dialogue sys- tems, also known as chatbots, have daily … infor- mation of the semantic units and achieve better performance such as text classification and relation …

Training neural response selection for task-oriented dialogue systems
M Henderson, I Vuli?, D Gerz, I Casanueva… – arXiv preprint arXiv …, 2019 – arxiv.org
… Abstract Despite their popularity in the chatbot liter- ature, retrieval-based models have had mod- est impact on task-oriented dialogue systems, with the main obstacle to their application be- ing the low-data regime of most task-oriented dialogue tasks …

Ensemble-based deep reinforcement learning for chatbots
H Cuayáhuitl, D Lee, S Ryu, Y Cho, S Choi, S Indurthi… – Neurocomputing, 2019 – Elsevier
… 10], [11], [12], [13], [14], [15], policy-based methods have been particularly applied to open-ended dialogue systems such as (chitchat) chatbots [5], [6], [16]. This is not surprising given the fact that task-oriented dialogue systems use finite action sets, while chatbot systems use …

Arabic Dialogue Act Recognition for Textual Chatbot Systems
A Joukhadar, H Saghergy, L Kweider… – Proceedings of The First …, 2019 – aclweb.org
… In general, chat bot systems can be composed of three basic components: Natural Language Un- derstanding (NLU … (5%) A dataset collected in a previous food order chatbot project (Shbib et … Se- quential short-text classification with recurrent and convolutional neural networks …

Automated scoring of chatbot responses in conversational dialogue
SK Yuwono, B Wu, LF D’Haro – … Workshop on Spoken Dialogue System …, 2019 – Springer
… forest (RF) is another supervised learning model which performs well for text classification [30 … paper, we have addressed the task of automated scoring of chatbot responses or … In: WOCHAT: workshop on chatbots and conversational agent technologiesGoogle Scholar. 3. Banchs …

Enriching Conversation Context in Retrieval-based Chatbots
AV Tahami, A Shakery – arXiv preprint arXiv:1911.02290, 2019 – arxiv.org
… In this paper, we introduced a new architecture for use in retrieval-based chat- bots … structurally sound model for the relations between contexts and responses, similar to work done previously on text classification [26 … Enriching Conversation Context in Retrieval-based Chatbots …

CONVERSATIONAL CHATBOT SYSTEM FOR STUDENT SUPPORT IN ADMINISTRATIVE EXAM INFORMATION
HA Rasheed, J Zenkert, C Weber, M Fathi – researchgate.net
… Traditional text classification methods include naïve Bayes, support vector machine (SVM), and k- nearest neighbour … Page 8. [3] S. Abdul-Kader and Dr. John, “Survey on Chatbot Design Techniques in … [8] D. Jurafsky and J. Martin, “Dialog Systems and Chatbots,” Speech and …

Convolutional neural encoder for the 7th dialogue system technology challenge
M Korpusik, J Glass – … of 2019 AAAI 7th Dialogue System …, 2019 – workshop.colips.org
… The 7th dialog system technology challenge. arXiv preprint. Zhang, X.; Zhao, J.; and LeCun, Y. 2015. Character-level convolutional networks for text classification … Multi-turn response selection for chatbots with deep attention matching network …

EMERGENCY PATIENT CARE SYSTEM USING CHATBOT
P Raj, R Murali Krishna, SM Krishna, KH Vardhan… – ijtre.com
… aid, offering a solution for simpler medical issues: these are all possible situations for chatbots to step … C. Multinomial Naive Bayes: Multinomial Naive Bayes is an algorithm for text classification and Natural language processing … An Approach to Enhance Chatbot Semantic Power …

Towards Emotion Intelligence in Neural Dialogue Systems
C Huang – 2019 – era.library.ualberta.ca
… 13 2.5 Neural Models for Text Classification … With the recent development of Deep Learning (DL), building open-domain social chatbots has received much more attention from the research community … social textual chatbot XiaoIce [104] states that a mature chatbot should be …

Survey on Out-Of-Domain Detection for Dialog Systems
YS Jeong, YM Kim – Journal of Convergence for Information …, 2019 – koreascience.or.kr
… Key Words : Dialog system, User utterance, Out-of-domain detection, Natural language understanding, Text classification … to intelligent conversational agent, which is also typically called as chatbot or dialog … direction is to construct a public dataset for the dialog system, and this …

CBET: design and evaluation of a domain-specific chatbot for mobile learning
Q Liu, J Huang, L Wu, K Zhu, S Ba – Universal Access in the Information …, 2019 – Springer
… To enhance the chatbots applied in mobile learning, we pro- pose an intelligent chatbot for the … eg, greetings, emo- tion, and humor into the embedded KB to make the chatbot more user … is conducted by learning with posi- tive data only and is applied in short text classification …

Evaluating and Enhancing the Robustness of Retrieval-Based Dialogue Systems with Adversarial Examples
J Li, C Tao, N Peng, W Wu, D Zhao, R Yan – CCF International Conference …, 2019 – Springer
… it to natural language processing (NLP) tasks, such as text classification [4], question … design adversarial example generation methods, which successfully fool retrieval-based dialog systems. We significantly improve the robustness of retrieval-based dialogue systems with our …

Deep reinforcement learning for chatbots using clustered actions and human-likeness rewards
H Cuayáhuitl, D Lee, S Ryu, S Choi… – … Joint Conference on …, 2019 – ieeexplore.ieee.org
… This is not surprising given the fact that task-oriented dialogue systems use finite action sets, while chatbot systems use infinite action sets. So far there is a preference for policy search methods for chatbots, but it is not clear whether they should be preferred because they face …

Improvement of Chatbots Semantics Using Wit. ai and Word Sequence Kernel: Education Chatbot as a Case Study.
AA Qaffas – International Journal of Modern Education & Computer …, 2019 – j.mecs-press.net
… Res., 5:361– 397, 2004. [10] H. Lodhi, N. Cristianini, J. Shawe-Taylor, and C. Watkins. Text classification using string kernel … How to cite this paper: Alaa A. Qaffas, ” Improvement of Chatbots Semantics Using Wit.ai and Word Sequence Kernel: Education Chatbot as a …

A dialogue-based annotation for activity recognition
T Mairittha, N Mairittha, S Inoue – Adjunct Proceedings of the 2019 ACM …, 2019 – dl.acm.org
… 3.1 A dialogue-based annotation Dialogue system, also known as a conversational agent, virtual agent, or chatbot [12]. It is the system that communicates with … Bag of tricks for efficient text classification. arXiv preprint arXiv:1607.01759 (2016) … From chatbots to dialog systems …

Evaluation and improvement of chatbot text classification data quality using plausible negative examples
K Kuksenok, A Martyniv – arXiv preprint arXiv:1906.01910, 2019 – arxiv.org
… uses plausible negative examples to perform actionable, model-agnostic evaluation of text classification as a … Commercial chatbot: per- formance evaluation, usability metrics and quality standards of embodied … A quality analy- sis of facebook messenger’s most popular chatbots …

Improving human-computer interaction in low-resource settings with text-to-phonetic data augmentation
A Stiff, P Serai, E Fosler-Lussier – ICASSP 2019-2019 IEEE …, 2019 – ieeexplore.ieee.org
… To address these problems, we propose a concept to improve perfor- mance in text classification tasks that use speech … Index Terms— Low-resource, spoken dialog systems, chatbot … use to test this concept, and what mo- tivates the work, is a virtual patient dialog system used to …

Typographic-Based Data Augmentation to Improve a Question Retrieval in Short Dialogue System
HS Nugraha, S Suyanto – 2019 International Seminar on …, 2019 – ieeexplore.ieee.org
… in a data set collected from a social media-based Indonesian chatbot, two sentences … Fig. 1. Flow chart of the proposed question retrieval system for short dialog system … 18] X. Zhang, J. Zhao, Y. LeCun, “Character-level convolutional networks for text classification,” In: Advances …

Data collection methods for building a free response training simulation
V Sharma, B Shpringer, SM Yang… – 2019 Systems and …, 2019 – ieeexplore.ieee.org
… used to train a variety of models to produce general conversational chatbot systems [20 … implemented for replacing a multiple- choice dialogue system with a text classification based free … L. Dybkjaer, & NO Bernson (Eds.), Advances in natural multimodal dialog systems New York …

Resilient chatbots: repair strategy preferences for conversational breakdowns
Z Ashktorab, M Jain, QV Liao, JD Weisz – … of the 2019 CHI Conference on …, 2019 – dl.acm.org
… extraction can be achieved through various methods for any kind of text classification algorithms [35]; thus … help users to rephrase; teach user how to interact with the chatbot; proactively making … help users to rephrase; teach user how to interact with chatbots; proactively making …

Fine-grained sentence functions for short-text conversation
W Bi, J Gao, X Liu, S Shi – arXiv preprint arXiv:1907.10302, 2019 – arxiv.org
… Research on dialogue systems or chatbots have studied to control the output responses with dif … 2003) find that many conversation breakdowns could be avoided if the chatbot can recognize … The first is a CNN-based encoder commonly used for text classification tasks (Kim, 2014 …

Exploring Social Bias in Chatbots using Stereotype Knowledge
N Lee, A Madotto, P Fung – Proceedings of the 2019 Workshop on …, 2019 – winlp.org
… In this work, we propose to use the NLI module to serve as the crucial interpreter that maps chatbot response to the … We hope our findings in this work can initiate more research to ensure fairness in chatbots … 2017. Measuring and mitigating unintended bias in text classification …

Online Communication with Natural Language
D Sathya, D Jagadeesan, P Betty – … International Conference on …, 2019 – ieeexplore.ieee.org
… Chatbots make college websites more efficient and approachable than earlier … The result of the proposed project lies in developing a conversational chatbot which can efficiently … The use of text classification has been very helpful for keyword matching with their simplicity and …

Proceedings of the First Workshop on NLP for Conversational AI
YN Chen, T Bedrax-Weiss, D Hakkani-Tur… – Proceedings of the First …, 2019 – aclweb.org
… Building a Production Model for Retrieval-Based Chatbots Kyle Swanson, Lili Yu, Christopher Fox … Evaluation and Improvement of Chatbot Text Classification Data Quality Using Plausible Negative Examples Kit … Do Neural Dialog Systems Use the Conversation History Effectively …

Deep learning for spoken dialogue systems: application to nutrition
MB Korpusik – 2019 – dspace.mit.edu
… partment of Defense (DoD) through the National Defense Science & Engineering Gradu- ate Fellowship (NDSEG) Program. 8 Page 9. Contents 1 Introduction 31 1.1 Dialogue Systems … 164 7.4 7th Dialogue System Technology Challenge (DSTC7) …

JAQ: a chatbot for foreign students
A Gellens, S Gustin, Y Deville – dial.uclouvain.be
… A chatbot (also called chatterbot, conversational agent or sometimes spoken dialogue system) is a computer program able to converse … Figure 2.1: Chatbot abstract view … When talking about chatbots, there can be a potential confusion with other types of systems, either because …

Reconstructing capsule networks for zero-shot intent classification
H Liu, X Zhang, L Fan, X Fu, Q Li, XM Wu… – Proceedings of the 2019 …, 2019 – aclweb.org
… With the advent of conversational AI, task- oriented spoken dialogue systems are becoming ubiquitous, eg, chatbots deployed on differen- t applications, or modules integrated in the pop- ular virtual personal assistants like Apple Siri or Microsoft Cortana (Chen et al., 2017) …

Sentence Similarity Techniques for Short vs Variable Length Text using Word Embeddings
D Shashavali, V Vishwjeet, R Kumar, G Mathur… – Computación y …, 2019 – cys.cic.ipn.mx
… Sentence similarity, word embeddings, natural language processing, sliding window, N-grams, text classification … In applications like Chatbots, the uses of sentence similarity include estimating the semantic … In Chatbot application, False Positive must be very less for better user …

SuperChat: dialogue generation by transfer learning from vision to language using two-dimensional word embedding
B Sun, L Yang, M Lin, C Young, J Dong… – Proceedings of the 1st …, 2019 – dl.acm.org
… KEYWORDS System Demo, Chatbot, Two-dimensional Word Embedding, Dia- logue Generation, Transfer Learning … Given an input sen- tence, the dialogue system outputs the response sentence in a nat … 11] has obtained state-of-the-art result for text classification on benchmark …

Intent Classification for Dialogue Utterances
J Schuurmans, F Frasincar – IEEE Intelligent Systems, 2019 – ieeexplore.ieee.org
… Hierarchical classification was first used for text classification by Koller and Sahami.15 They … There is the Chatbot Corpus on Travel Scheduling, and the StackExchange Corpus on … A. Wroblewska, “Multi- intent hierarchical natural language understanding for chatbots,” in Proc …

Implementation and evaluation of a shopping assistance chatbot in an e-commerce case
T Böger – 2019 – run.unl.pt
… This can be for example information retrieval from a knowledge base, true machine translation, or a dialogue system (Chowdhury, 2005 … At first, a definition of the term chatbot is given to then explain different types of bots. Chatbots in the literature share several different names …

From user feedback to requirements using chatbots
D Horváth – 2019 – dspace.library.uu.nl
… Furthermore, chatbots can help in guiding the users through the bug reporting process by controlling the flow of the conversation and using follow-up … After the evaluation the subjects were asked to rate the chatbot using a usab- ility scale, and were interviewed to find out their …

Say what i want: Towards the dark side of neural dialogue models
H Liu, T Derr, Z Liu, J Tang – arXiv preprint arXiv:1909.06044, 2019 – arxiv.org
… people with ulterior motives may take advantage of this weakness of the chatbots to guide them … Text classification problems are studied in [30, 31]; sentiment analysis is involved in [32]; grammar error … It is unachievable in reality when hackers try to manipulate a chatbot service …

ANA at SemEval-2019 Task 3: Contextual Emotion detection in Conversations through hierarchical LSTMs and BERT
C Huang, A Trabelsi, OR Zaïane – arXiv preprint arXiv:1904.00132, 2019 – arxiv.org
… show that, in this task, our HRCLE outperforms the most recent state-of- the-art text classification framework: BERT … In such cases, a user is conversing with an automatic chatbot. Empowering the chat- bot with the ability to detect the user’s emotion is a step forward towards the …

Speech Command Classification System for Sinhala Language based on Automatic Speech Recognition
T Dinushika, L Kavmini… – … Conference on Asian …, 2019 – ieeexplore.ieee.org
… It detects the intent of a Sinhala utterance using ASR followed by text classification … this speech command classification system and develop a speech dialog system for the … Bayes Algorithm And Logistic Regression For Intent Classification In Chatbot,” International Conference …

Short Text Conversation Based on Deep Neural Network and Analysis on Evaluation Measures
HE Cherng, CH Chang – arXiv preprint arXiv:1907.03070, 2019 – arxiv.org
… methods that evaluate the quality and structure of dialogue between a chatbot and a … With such measures, the quality of chatbots could be evaluated automatically and efficiently … Retrieval-based dialogue system requires a knowledge base with large amount of question-answer …

FastText-Based Intent Detection for Inflected Languages
K Balodis, D Deksne – Information, 2019 – mdpi.com
… system, the specific nature of intent detection that separates it from other text classification tasks is … a large amount of training data, which is not the typical case for chatbots … The chatbot dataset contains users’ questions from a Telegram chatbot that answers questions related to …

Multilingual Dialogue Generation with Shared-Private Memory
C Chen, L Qiu, Z Fu, J Liu, R Yan – CCF International Conference on …, 2019 – Springer
… systems accomplish a specific task and non-task-oriented dialogue systems are designed to chat in open domain as chatbots [1]. In … [23] propose an adversarial multi-task learning framework for text classification … Yan, R.: Chitty-Chitty-Chat Bot: deep learning for conversational AI …

Rhetorical Agreement: Maintaining Cohesive Conversations
B Galitsky – Developing Enterprise Chatbots, 2019 – Springer
… In recent years, development of chatbots answering questions, facilitation discussion, managing dialogues … of how a question answering, dialog management, recommendation or chatbot system can … the last two decade, research in the field of dialogue systems has experienced …

Passive Diagnosis of Mental Health Disorders Incorporating an Empathic Dialogue System
F Delahunty, M Arcan, R Johansson – 2019 – thesiscommons.org
… 3Commonly known as chatbots … par- ticipants from the public domain who had a short conversation with our proposed empathic dialogue system … To evaluate Hypothesis 2, we randomly allocated recruited participants to have conversations with one of two dialogue systems …

Classification As Decoder: Trading Flexibility For Control In Neural Dialogue
S Shleifer, M Chablani, N Katariya, A Kannan… – arXiv preprint arXiv …, 2019 – arxiv.org
… Many production chatbots check each word in a generated utterance against a blacklist of curse words, but this fails … The closest work to ours is Wan & Chen (2018)’s AirBNB customer service chatbot, which also uses a … Universal language model fine-tuning for text classification …

A multi-task hierarchical approach for intent detection and slot filling
M Firdaus, A Kumar, A Ekbal… – Knowledge-Based Systems, 2019 – Elsevier
… Also, by handling intent and slot together, we can build an end-to-end natural language understanding (NLU) module for any task-oriented chatbot … 2. Related work. SLU is an integral part of every dialogue system. Dialogue systems are mostly like knowledge-based systems …

Exploring machine learning and deep learning frameworks for task-oriented dialogue act classification
T Saha, S Srivastava, M Firdaus, S Saha… – … Joint Conference on …, 2019 – ieeexplore.ieee.org
… Components of Spoken Dialogue System (SDS) such as Natu- ral Language Understanding (NLU) and … Applications such as online chat-bots that include the Problem Solving Agent, Conversational Agent … set is proposed which is more appropriate for building a chat-bot system …

Trends in Deep-neural-network-based Dialogue Systems
OW Kwon, TG Hong, JX Huang, YH Roh… – Electronics and …, 2019 – koreascience.or.kr
… 499–507. [15] JX Huang et al., “Improve the Chatbot Performance for … [32] DSTC8, “The Eighth Dialog System Technology Challenge,” https:// sites.google.com/dstc.community/dstc8 … 135-139. [35] JY Lee and F. Dernoncourt, “Sequential Short-Text Classification with Recurrent …

Classification as Decoder: Trading Flexibility for Control in Multi Domain Dialogue
S Shleifer, M Chablani, N Katariya, A Kannan… – 2019 – openreview.net
… Many production chatbots check each word in a generated utterance against a blacklist of curse words, but this fails … The closest work to ours is Wan & Chen (2018)’s AirBNB customer service chatbot, which also uses a … Universal language model fine-tuning for text classification …

Towards the Learning, Perception, and Effectiveness of Teachable Conversational Agents
N Chhibber – 2019 – uwspace.uwaterloo.ca
… We also provide an evaluation of how crowdworkers perform a text classification before and after interacting with a teachable agent … 19 3.3.1 Why a Conversational Interface? . . . . 19 3.3.2 Dialog System . . . . . 20 …

Unified language model pre-training for natural language understanding and generation
L Dong, N Yang, W Wang, F Wei, X Liu… – Advances in Neural …, 2019 – papers.nips.cc
… mixed precision training. 2.5 Fine-tuning on Downstream NLU and NLG Tasks For NLU tasks, we fine-tune UNILM as a bidirectional Transformer encoder, like BERT. Take text classification as an example. We use the encoding …

Enhancing Out-Of-Domain Utterance Detection with Data Augmentation Based on Word Embeddings
Y Feng, J Lin – arXiv preprint arXiv:1911.10439, 2019 – arxiv.org
… utterances for conversational assistant platforms is even more challenging than building chatbots for one … Whereas building a domain-specific chatbot can rely on collecting OOD samples iteratively … By using a CNN-based model for text classification, we find a positive correlation …

Deep neural architecture with character embedding for semantic frame detection
FZ Daha, S Hewavitharana – 2019 IEEE 13th International …, 2019 – ieeexplore.ieee.org
… has many applications and has been useful for systems like chat bots [1], question … Ilievski, C. Musat, A. Hossmann, and M. Baeriswyl, “Goal- oriented chatbot dialog management … Zhao, and Y. LeCun, “Character-level convolutional networks for text classification,” in Proceedings …

Optimizing the Design and Cost for Crowdsourced Conversational Utterances
P Liu, J Xiao, T Liu, DF Glas – 2019 – dylanglas.com
… A requester is interested in collecting 100 utterance variants for training a chatbot, and he … layer perceptron (MLP), which has been demonstrated to provide competitive results on text classification task [2 … Data Collection for a Production Dialogue System : A Clinc Perspective …

Learning Multi-Party Turn-Taking Models from Dialogue Logs
MG de Bayser, P Cavalin, C Pinhanez… – arXiv preprint arXiv …, 2019 – arxiv.org
… The expert chatbots are not only able to give answers related to investments but can also … machine interactions and is more topic-oriented in a way that each chatbot interacts only … AC-CNN) presented herein consists of a standard model used for text classification adapted for the …

Comparing the Performance of Feature Representations for the Categorization of the Easy-to-Read Variety vs Standard Language
M Santini, B Danielsson, A Jönsson – … of the 22nd Nordic Conference on …, 2019 – aclweb.org
… retrieval (eg for the retrieval of easy-to-read or patient-friendly medical information) and deep learning-based dialogue systems (eg customized chatbots for expert … items in the form of BoWs represent the topic(s) of a text and are nor- mally used for topical text classification …

Modeling both Context-and Speaker-Sensitive Dependence for Emotion Detection in Multi-speaker Conversations.
D Zhang, L Wu, C Sun, S Li, Q Zhu, G Zhou – IJCAI, 2019 – ijcai.org
… et al., 2019], and intelligent systems like smart homes and chatbots [Young et al., 2018] … GCN is also exported to sev- eral NLP tasks such as semantic role labeling [Marcheggiani and Titov, 2017], text classification [Yao et al., 2019] and ma- chine translation [Bastings et al., 2017 …

An Evaluation of Chinese Human-Computer Dialogue Technology
Z Zhao, W Zhang, W Che, Z Chen, Y Zhang – Data Intelligence, 2019 – MIT Press
… A similar evaluation based on English corpora is the 6th Dialog System Technology Challenges (DSTC6 … set which contains 31 user intents that appear frequently in general-purpose chatbots … This also indicates that many of the current models for text classification tasks have …

TKUIM at NTCIR-14 STC-3 CECG task
S Wei, C Cheng, Y Guang-zhong-yi Cao, CW Chiang… – research.nii.ac.jp
… Most chatbots are based on generative models, which can be improved under the Seq2Seq … In the following we will introduce several techniques related to the generative chatbot used in this … Kim (2014) mentioned a series of well-performing CNN text classification experiments …

Wordnet as a Relational Semantic Dictionary Built on Corpus Data
M Piasecki, A Dziob – CLARIN, 2019 – videolectures.net
Page 1. CLARIN-PL Wordnet as a Relational Semantic Dictionary Built on Corpus Data Maciej Piasecki, Agnieszka Dziob Wroc?aw University of Science and Technology G4.19 Research Group maciej.piasecki@pwr.edu.pl 2019-09-03 Page 2. Plan …

Personalizing dialogue agents via meta-learning
A Madotto, Z Lin, CS Wu, P Fung – … of the 57th Annual Meeting of the …, 2019 – aclweb.org
… a growing interest in learning personal- ized chit-chat dialogue agents for making chat- bots more consistent … 2018), machine translation for low resource language (Gu et al., 2018), and for text classification (Yu et … Learning from dialogue after deployment: Feed yourself, chatbot …

Out-of-Domain Detection for Low-Resource Text Classification Tasks
M Tan, Y Yu, H Wang, D Wang, S Potdar… – arXiv preprint arXiv …, 2019 – arxiv.org
… son Assistant1. For example, Table 1 shows some of the utterances a chat-bot builder provided for training … At- tentive task-agnostic meta-learning for few-shot text classification … em- bedding using only in-domain sentences for out-of- domain sentence detection in dialog systems …

Intelligent Chatbot for Requirements Elicitation and Classification
CSRK Surana, DB Gupta… – 2019 4th International …, 2019 – ieeexplore.ieee.org
… In this paper, automation of requirements elicitation and classification using conversational chatbot technology and text classification algorithms has been presented … Programming challenges of chatbot: Current and future prospective … Build Better Chatbots. Apress …

Bootstrapping conversational agents with weak supervision
N Mallinar, A Shah, R Ugrani, A Gupta… – Proceedings of the AAAI …, 2019 – aaai.org
… also identified areas for future work to improve this new approach for developing chatbots, as well … tion tasks explored in Ratner et al.’s work, we focus on text classification, where the … Moreover, chatbot development of- ten requires revising the intents as end users’ behaviors or …

A repository of conversational datasets
M Henderson, P Budzianowski, I Casanueva… – arXiv preprint arXiv …, 2019 – arxiv.org
… 1 Introduction Dialogue systems, sometimes referred to as con- versational systems or conversational agents, are useful in a wide array of applications … However, collecting data to train data-driven dialogue systems has proven notori- ously difficult …

An attentive survey of attention models
S Chaudhari, G Polatkan, R Ramanath… – arXiv preprint arXiv …, 2019 – arxiv.org
… Question Answering, Sentiment Analysis, Part-of-Speech tagging, Constituency Parsing and Dialogue Systems … In contrast, for tasks such as text classification and rec- ommendation, input is a … 4.2 Memory Networks Applications like question answering and chat bots require the …

The eighth dialog system technology challenge
S Kim, M Galley, C Gunasekara, S Lee… – arXiv preprint arXiv …, 2019 – arxiv.org
… with limited in-domain data, for example, when modeling user responses for a task-oriented chatbot on a … video description models into a single end-to-end differentiable network to build scene-aware dialog systems … ConvLab: Multi-domain end-to-end dialog system platform …

Learning to Customize Language Model for Generation-based dialog systems
Y Song, Z Liu, W Bi, R Yan, M Zhang – arXiv preprint arXiv:1910.14326, 2019 – arxiv.org
… Experiment results show that our algo- rithm excels all the baselines in terms of personality, quality, and diversity measurement. Introduction Building personalized chatbots that can interact with differ- ent users with according content and language styles (Li et al …

Text Generation for Imbalanced Text Classification
S Akkaradamrongrat, P Kachamas… – … Joint Conference on …, 2019 – ieeexplore.ieee.org
… Text generation can be applied to many fields such as chatbots, lyric generation, question … [7], applied LSTM to generate natural language for spoken dialogue systems by giving … of this research is to find the method which can solve the problem of imbalanced text classification …

Cross-Domain Training for Goal-Oriented Conversational Agents
AM Bodîrl?u, S Budulan, T Rebedea – Proceedings of the International …, 2019 – aclweb.org
… domain to boost the training and testing performance of a machine learning chatbot model on a … (2018) use the transfer learning mechanism for chatbots employing neural … based tuning, which achieves a mean GLUE score of 80.0 on several text classification tasks, compared to …

Unsupervised dialogue intent detection via hierarchical topic model
A Popov, V Bulatov, D Polyudova… – Proceedings of the …, 2019 – aclweb.org
… One of the challenges during a task- oriented chatbot development is the scarce availability of … More universal and robust dialogue systems should work without any supervision or defined rules … Section two describes popular approaches to an unsuper- vised text classification …

A study of incorrect paraphrases in crowdsourced user utterances
MA Yaghoub-Zadeh-Fard, B Benatallah… – Proceedings of the …, 2019 – aclweb.org
… Also known as dialogue systems, virtual assistants, chatbots or simply bots (Campagna et al., 2017; Su et al., 2017 … Moreover, it is feasible to automatically generate pairs of questions and answers by mining datasets in the fields of Question Answering and dialog systems …

One time of interaction may not be enough: Go deep with an interaction-over-interaction network for response selection in dialogues
C Tao, W Wu, C Xu, W Hu, D Zhao, R Yan – … of the 57th Annual Meeting of …, 2019 – aclweb.org
… study the prob- lem of multi-turn response selection for retrieval- based dialogue systems where the … fluency and diversity, and thus have been widely applied in commercial chatbots such as … vi- sion, they have proven effective in a few tasks such as text classification (Conneau et …

A Study on Named Entity Recognition for Effective Dialogue Information Prediction
M Go, H Kim, H Lim, Y Lee, M Jee… – Journal of Broadcast …, 2019 – koreascience.or.kr
… Oriented Dialogue System)? ??? ?? ?? ??? (ChatBot)?? ?? ? ?? … K. Lee and Y. Kim, “A Chatter Bot for a Task-Oriented Dialogue System”, KIPS Transactions … J. Zhao and Y. LeCun, “Character-level Convolutional Networks for Text Classification”, Advances in …

Incremental Improvement of a Question Answering System by Re-ranking Answer Candidates using Machine Learning
M Barz, D Sonntag – arXiv preprint arXiv:1908.10149, 2019 – arxiv.org
… EVORUS learns to select answers from multiple chatbots via crowdsourcing [11] … Text classification for tensorflow embedding is done using TensorFlow with an implementation of the StarSpace … The logs of a deployed chatbot, that contain actual user queries, can be efficiently …

Open Domain Conversational Chatbot
V Deshmukh, SJ Nirmala – International Conference on Information …, 2019 – Springer
… There is a huge range of chatbots identified based on the learning capacity and the … Gaussian assumes the distribution as continuous and multinomial used in text classification … N., Bao, J., Chen, P., Zhou, M.: DocBot: an information retrieval approach for chatbot engines using …

Introducing MANtIS: a novel multi-domain information seeking dialogues dataset
G Penha, A Balan, C Hauff – arXiv preprint arXiv:1912.04639, 2019 – arxiv.org
… 4 A generic dialogue system is composed of the following: natural language understand- ing ? dialogue … Domain specific dialogue systems do not generalize to new/unseen information needs … multiple intents, this is a multi-label and multi-class text classification problem, and …

Deep Reinforcement Learning for Text and Speech
U Kamath, J Liu, J Whitaker – Deep Learning for NLP and Speech …, 2019 – Springer
… systems. In the next sections, we provide a survey of different DRL methods for information extraction, text classification, dialogue systems, text summarization, machine translation, and natural language generation. Many of …

Standardization of Robot Instruction Elements Based on Conditional Random Fields and Word Embedding
H Wang, Z Zhang, J Ren, T Liu – Journal of Harbin Institute of …, 2019 – hit.alljournals.cn
… The dialog system [2-3] for human robot interaction through natural spoken language is … about destination, transportation, accommodation, etc., and also different from ordinary chatbots which usually … in many NLP tasks such as language translation [14] and text classification [15 …

Optimising user experience with: conversational Interfaces
AMG Costa – 2019 – recipp.ipp.pt
… Keywords: Conversational Interfaces, Deep Learning, Natural Language Processing, Chat- bots, User Experience, CRM. I … can be specifically formatted if you are communicating with this chatbot via the Messenger … The chatbots come in two flavors, rule-based and AI bots …

Personalized dialogue generation with diversified traits
Y Zheng, G Chen, M Huang, S Liu, X Zhu – arXiv preprint arXiv …, 2019 – arxiv.org
… WORK It has been demonstrated that personality is vital for building a human-like dialogue system [15, 39 … Therefore, “Big Five” is not suitable for building large-scale personalized dialogue systems, particularly with data-driven neural models … 32], in which a chat- bot is endowed …

The speech interface as an attack surface: an overview
M Bispham, I Agrafiotis, M Goldsmith – International Journal On …, 2019 – ora.ox.ac.uk
… Current task-based dialogue systems have some similarity with chatbots in that they are … controlled digital assistants include, in addition to the generic speech dialogue system components, an … [53] adversarial learning natural language understanding in text classification system …

Transfer Hierarchical Attention Network for Generative Dialog System
X Zhang, Q Yang – International Journal of Automation and Computing, 2019 – Springer
… The Chit-chat dialog system is a promising natural language processing (NLP) technology which aims to en- able computers to chat with human through natural lan- guage. Traditional chit-chat dialog systems are built by hand-crafted rules or directly selecting a human writing …

Chatterbox: Conversational interfaces for microtask crowdsourcing
P Mavridis, O Huang, S Qiu, U Gadiraju… – Proceedings of the 27th …, 2019 – dl.acm.org
… Examples of such microtasks include audio/text transcription, image/text classification, and information finding … [29] investigated the use of chatbot for styling … crowdsourcing addressed the integration of crowd work platforms with text-messaging and chatbots systems, mostly to …

ProductQnA: Answering user questions on e-commerce product pages
A Kulkarni, K Mehta, S Garg, V Bansal… – … Proceedings of The …, 2019 – dl.acm.org
… KEYWORDS question answering; deep learning; chatbot; e-commerce ACM Reference Format: Ashish Kulkarni … summarization, and re- cently, automatic question answering [17, 20] and chatbots [22] … to com- plex deep learning models such as LSTM for text classification [5] as …

A Comparative Study of Classical and Deep Classifiers for Textual Addressee Detection in Human-Human-Machine Conversations
O Akhtiamov, D Fedotov, W Minker – International Conference on Speech …, 2019 – Springer
… Text classification Speaking style Human-computer interaction Spoken dialogue system. Download conference … This capability is extremely useful for SDSs, such as personal assistants, social robots, and chat bots, and may essentially improve their performance …

Dialogue quality and nugget detection for short text conversation (STC-3) based on hierarchical multi-stack model with memory enhance structure
HE Cherng, CH Chang – NTCIR14. p. to appear, 2019 – research.nii.ac.jp
… Automatic question-answering and dialog systems are important applications in enterprise customer services … plenty of time and human resources, and provide a 24-hour chatbot to answer … proposed in NTCIR-12 as the first step toward natural language conversation for chatbots …

Build it break it fix it for dialogue safety: Robustness from adversarial human attack
E Dinan, S Humeau, B Chintagunta… – arXiv preprint arXiv …, 2019 – arxiv.org
… fo- rums (Galán-Garc?a et al., 2016), and the de- ployment of chatbots in the … Adversarial attacks on the Tay chatbot led to the developers shutting down the system (Wolf … Nevertheless algo- rithmic approaches have been attempted, for ex- ample in text classification (Ebrahimi et …

MIDAS: A dialog act annotation scheme for open domain human machine spoken conversations
D Yu, Z Yu – arXiv preprint arXiv:1908.10023, 2019 – arxiv.org
… We publish the annotated human-machine chatbot corpus that has 24,000 utterances … While the dialog system may not need to change the topic if the utterance is User2A, but will … Building on previous work on text classification, we use an encoder-decoder model with two major …

User intent prediction in information-seeking conversations
C Qu, L Yang, WB Croft, Y Zhang, JR Trippas… – Proceedings of the 2019 …, 2019 – dl.acm.org
… and character level [32]. These new deep learn- ing techniques have been applied in medical dialog systems [4]. In this paper, we focus on user intent prediction in information- seeking conversations. This specific utterance …

# MeTooMaastricht: Building a chatbot to assist survivors of sexual harassment
T Bauer, E Devrim, M Glazunov, WL Jaramillo… – … Conference on Machine …, 2019 – Springer
… classified as rule-based (scripted) or end-to-end (usually based on deep learning) chatbots … doctoral fellow at Brandeis University, in building the dialogue flow of the chatbot implemented in … T., Tang, X.: A comparative study of TF* IDF, LSI and multi-words for text classification …

Computational linguistics: Introduction to the thematic issue
A Gelbukh – Computación y Sistemas, 2019 – cys.cic.ipn.mx
… The typical tasks of natural language processing include machine translation, text classification, text summarization … They write: Retrieval- based dialogue systems converse with humans by ranking candidate … write: In goal-oriented conversational agents like Chatbots, finding the …

Samvaadhana: A Telugu Dialogue System in Hospital Domain
SR Duggenpudi, KSS Varma, R Mamidi – … of the 2nd Workshop on Deep …, 2019 – aclweb.org
… 2017. A survey on dialogue systems: Recent advances and new frontiers. CoRR, abs/1711.01731. Grace Chung. 2004 … 2019. Spoken dialogue system using recognition of user’s feedback for rhythmic dialogue. Page 8. 241 … Gotora. 2017. A neural-network based chat bot …

Data Science and Conversational Interfaces: A New Revolution in Digital Business
D Griol, Z Callejas – Data Science and Digital Business, 2019 – Springer
… In a survey from 2016, 80% of business leaders stated that they either “already used or planned to use Chatbots by 2020” when asked which … Text classification from labeled and unlabeled documents using EM … Proposal of open-ended dialog system based on topic maps …

Persona-aware Dialogue Generation with Enriched Profile
Y Zheng, G Chen, M Huang, S Liu, X Zhu – alborz-geramifard.com
… (2018), in which a chatbot is endowed with a structured profile … Challenges in building intelligent open-domain dialog systems. arXiv preprint arXiv:1905.05709 … Recurrent convolutional neural networks for text classification. In AAAI, pages 2267–2273 …

Your instruction may be crisp, but not clear to me!
P Pramanick, C Sarkar… – 2019 28th IEEE …, 2019 – ieeexplore.ieee.org
… A robot deployed in our surrounding can utilize such a chatbot to derive the human intention and set a goal … Most of the existing chatbots are trained with query- response pairs and a given query is classified to such … We model the intent predication as a text classification problem …

Personalizing dialogue agents via meta-learning
Z Lin, A Madotto, CS Wu, P Fung – arXiv preprint arXiv:1905.10033, 2019 – arxiv.org
… a growing interest in learning personal- ized chit-chat dialogue agents for making chat- bots more consistent … 2018), machine translation for low resource language (Gu et al., 2018), and for text classification (Yu et … Learning from dialogue after deployment: Feed yourself, chatbot …

Evaluating and enhancing the robustness of dialogue systems: A case study on a negotiation agent
M Cheng, W Wei, CJ Hsieh – Proceedings of the 2019 Conference of the …, 2019 – aclweb.org
… on crafting adversarial agents instead of adversarial examples in an interac- tive dialogue system … that our algorithm can be generalized to other goal-oriented dialogue systems by designing a … In the negotiation chatbot setting, agents first chat using natural language and then …

Knowledge Creation Model for Emotion Based Response Generation for AI
UK Premasundera, MC Farook – 2019 19th International …, 2019 – ieeexplore.ieee.org
… Zhou and colleagues [21] at Tsinghua University in Beijing have developed a chatbot that can evaluate the … 25] verified that creating a model combining these two techniques will provide an accurate model for text classification … [22] A. Pardes, “The Emotional Chatbots Are Here …

MOLI: Smart Conversation Agent for Mobile Customer Service
G Zhao, J Zhao, Y Li, C Alt, R Schwarzenberg… – Information, 2019 – mdpi.com
… we introduce a conversational system, MOLI (the name of our dialogue system), to solve … Recent advances in dialog systems have led to successful applications in domains such as … Developing a dialog system for technical customer support presents additional challenges due to …

Listening between the lines: Learning personal attributes from conversations
A Tigunova, A Yates, P Mirza, G Weikum – The World Wide Web …, 2019 – dl.acm.org
… knowledge for personalization in downstream applications such as Web-based chatbots and agents … General-purpose chatbot-like agents show decent performance in benchmarks (eg, [13, 20, 37 … In this model futter is implemented with a text classification CNN [18] with a ReLU …

A topic-driven language model for learning to generate diverse sentences
C Gao, J Ren – Neurocomputing, 2019 – Elsevier
… Current generative applications include text summarization [1], question generation [2], response generation in dialogue systems [3], [4 … 7] generate the responses of the chatbot based on … In the human-machine dialogue system, we can get a topic distribution of an input sentence …

“I think it might help if we multiply, and not add”: Detecting Indirectness in Conversation
P Goel, Y Matsuyama, M Madaio, J Cassell – … on Spoken Dialogue System …, 2019 – Springer
… Tutoring Agent or a general-purpose socially-aware spoken dialogue system [30] that can … in interpersonal communication, incorporating its detection in spoken dialogue systems may ultimately … Lee JY, Dernoncourt F (2016) Sequential short-text classification with recurrent and …

Conversation Model Fine-Tuning for Classifying Client Utterances in Counseling Dialogues
S Park, D Kim, A Oh – arXiv preprint arXiv:1904.00350, 2019 – arxiv.org
… Mobile-based psychotherapy programs (Mantani et al., 2017), fully automated chatbots (Ly et al., 2017; Fitzpatrick et al., 2017), and intervention through smart devices (Torrado et al., 2017) are examples … (3) CNN for text classification (Kim, 2014) …

Semantic representations for under-resourced languages
J Mazarura, A de Waal, P de Villiers – … of the South African Institute of …, 2019 – dl.acm.org
… collaborative filtering, aspect-based sentiment analy- sis, intent classification for chatbots and machine … filtering [16, 33], aspect-based sentiment analysis [6] and text classification [24] … Intent classification, which is used in task-driven dialogue systems and sentiment classification …

Multiple Generative Models Ensemble for Knowledge-Driven Proactive Human-Computer Dialogue Agent
Z Dai, W Liu, G Zhan – arXiv preprint arXiv:1907.03590, 2019 – arxiv.org
… Generative method for building conversation chatbots has attracted increasing interest due to its great … For a knowledge-driven dialogue chatbot, an important ability is to reuse the … Eda: Easy data augmentation techniques for boosting performance on text classification tasks …

Multimodal open-domain conversations with robotic platforms
K Jokinen, G Wilcock – Multimodal Behavior Analysis in the Wild, 2019 – Elsevier
… Embodied conversational agents, chatbots, Siri, Amazon Alexa, Google Home, etc … of articles in Wikipedia, this can reasonably be called an open-domain dialog system … Conventional topic modeling and text classification have been mainly focused on static documents (ie …

Discourse-Level Dialogue Management
B Galitsky – Developing Enterprise Chatbots, 2019 – Springer
… 11.3.8 Evaluation: Information Access Efficiency in Chatbots Versus Search Engines … Twelve users (author’s colleagues) asked the chatbot 15–20 questions reflecting their financial situations … The structure of comparison of search efficiency for the chat bot vs the search engine is …

Recommendation as a communication game: Self-supervised bot-play for goal-oriented dialogue
D Kang, A Balakrishnan, P Shah, P Crook… – arXiv preprint arXiv …, 2019 – arxiv.org
… tifiable common goal. We leverage the dataset to develop an end-to-end dialogue system that can simultaneously converse and recommend. Models are … must accept it. A chatbot is then trained to play the expert in the game. (1) We …

An Architecture for Dynamic Conversational Agents for Citizen Participation and Ideation
S Ahmed – researchgate.net
… understanding components of Chatbots. The dialog management component of a Chatbot is traditionally implemented as a finite state machine with prompts repre … Neural netowrk approaches have been used extensively for text classification. This is …

Multi-attending Memory Network for Modeling Multi-turn Dialogue
J Ren, L Yang, C Zuo, W Kong, X Ma – Proceedings of the 2019 3rd High …, 2019 – dl.acm.org
… been widely applied, such as question answering [14][15][19], language modeling [15], text classification [16], and … dataset [9], which was collected or generated from a typical task-oriented multi-turn dialogue system … A Sequence to Sequence and Rerank based Chatbot Engine …

BoFGAN: Towards A New Structure of Backward-or-Forward Generative Adversarial Nets
MKS Chen, X Lin, C Wei, R Yan – The World Wide Web Conference, 2019 – dl.acm.org
… Processing, and it is also a critical step in many applications, such as machine translation [1, 33], dialogue systems [36, 38 … Deep learning methods have been widely used in the task of text classification and achieves good performance, especially using the convolutional neural …

Collaborative creation and training of social bots in learning communities
AT Neumann, P de Lange… – 2019 IEEE 5th …, 2019 – ieeexplore.ieee.org
… 6https://github.com/dennybritz/cnn-text-classification-tf 7https://github.com/karpathy/char-rnn 15 … A. Nguyen, J. Pineau, and Y. Bengio, “A deep reinforcement learning chatbot,” CoRR, vol … and O. D?az, “A quality analysis of facebook messenger’s most popular chatbots,” in SAC’18 …

Natural Language Processing, Understanding, and Generation
A Singh, K Ramasubramanian, S Shivam – Building an Enterprise Chatbot, 2019 – Springer
… Some chatbots are heavy on generative responses, and others are built for retrieving information and fitting it in … Since this book is about building an enterprise chatbot, we will focus more on the … fasText is a specialized library for learning word embeddings and text classification …

Curriculum Learning Strategies for IR: An Empirical Study on Conversation Response Ranking
G Penha, C Hauff – arXiv preprint arXiv:1912.08555, 2019 – arxiv.org
Page 1. Curriculum Learning Strategies for IR An Empirical Study on Conversation Response Ranking Gustavo Penha and Claudia Hauff TU Delft {g.penha-1,c.hauff}@ tudelft.nl Abstract. Neural ranking models are traditionally …

Chatbot Application on Cryptocurrency
Q Xie, D Tan, T Zhu, Q Zhang, S Xiao… – … IEEE Conference on …, 2019 – ieeexplore.ieee.org
… Chatbot knows the answer only because the input is in the associated pattern. Similarly, chatbots respond to anything related to the associated patterns, but it cannot go beyond the associated pattern … Naive Bayes is the classic algorithm for text classification …

Emotional State Estimation by Dialogue History and Sentence Distributed Representation
K Matsumoto, M Sasayama… – 2019 IEEE 6th …, 2019 – ieeexplore.ieee.org
… H.-Y., The Design and Implementation of XiaoIce, an Empathetic Social Chatbot, https://arxiv.org … [12] Kase, T., Nose, T., Chiba, Y., Ito, A. Evaluation of dialogue system using emotional … E., BojanowsNi, P., MiNolov, T. Bag of TricNs for Efficient Text Classification …

Deep learning for database mapping and asking clarification questions in dialogue systems
M Korpusik, J Glass – IEEE/ACM Transactions on Audio …, 2019 – ieeexplore.ieee.org
… recent work has shown improvements using deep CNN models for text classification [15]–[17 … showed that deep RL models enabled chatbots to generate more diverse, informative, and … Other work leveraged RL to construct a personalized dialogue system for a coffee-ordering …

A Novel E-mail Reply Approach for E-mail Management System
YE Feng – 2019 – orapp.aut.ac.nz
… level. Like Chatbots, it can predict the next sentence based on the last sentence … not limited to long text prediction but are suited for short-text prediction scenarios, such as Chatbots used in automatic reply in a chat room. 2. Using tags to improve the training corpus …

The Impact of Toxic Replies on Twitter Conversations
N Salehabadi – 2019 – rc.library.uta.edu
… Natural Language Processing (NLP) Techniques are used in Question-Answering, Chatbots and dialog systems. Generative models are state of the art for generating text, and … at al., 2018). In another work, (Zhang et al., 2018) created a dialogue system, where human …

Adversarial training in affective computing and sentiment analysis: Recent advances and perspectives
J Han, Z Zhang, B Schuller – IEEE Computational Intelligence …, 2019 – ieeexplore.ieee.org
Page 1. 68 IEEE ComputatIonal IntEllIgEnCE magazInE | may 2019 1556-603x/19© 2019IEEE Review Article Abstract over the past few years, adversar- ial training has become an extremely active research topic and has been …

A multi-encoder neural conversation model
D Ren, Y Cai, X Lei, J Xu, Q Li, H Leung – Neurocomputing, 2019 – Elsevier
… domain setting [9]. Zhu et al. [18] propose a fully data-driven generative dialogue system that is capable of generating responses based on input message and related knowledge base. To generate arbitrary number of answer …

Release strategies and the social impacts of language models
I Solaiman, M Brundage, J Clark, A Askell… – arXiv preprint arXiv …, 2019 – arxiv.org
… tions of chatbots alongside its latest release [19]. Finally, AI company AI21 recently announced work … lation systems, dialogue systems), often with less data and computing power than would be … A canonical example is Microsoft’s “Tay” chatbot, a Twitter bot that replied based on …

Ambient Assisted Living with Deep Learning
E Merdivan – 2019 – tel.archives-ouvertes.fr
… important components: improving activity recognition, addressing privacy concerns and developing intelligent dialogue systems for AAL systems, with an emphasis on a framework which is flexible and scalable for real-world applications. Page 20. Chapter 1. Introduction 3 …

A dynamic speaker model for conversational interactions
H Cheng, H Fang, M Ostendorf – Proceedings of the 2019 Conference of …, 2019 – aclweb.org
Page 1. Proceedings of NAACL-HLT 2019, pages 2772–2785 Minneapolis, Minnesota, June 2 – June 7, 2019. c 2019 Association for Computational Linguistics 2772 A Dynamic Speaker Model for Conversational Interactions …

Unsupervised Text Representation Learning with Interactive Language
H Cheng – 2019 – digital.lib.washington.edu
… neural-based spoken dialogue systems [36, 37] … Page 24. 13 tion scenario takes place between a human user and a dialogue system powered by conversational … Page 25. 14 (aka chatbots) have been developed for entertainment, companionship and education purpose …

BERT for Open-Domain Conversation Modeling
X Zhao, Y Zhang, W Guo, X Yuan – 2019 IEEE 5th International …, 2019 – ieeexplore.ieee.org
… It has been successfully applied to question answering, text classification, text augmentation, text … Bengio, Y., Courville, A., Pineau, J.: Building end-to-end dialogue systems using generative … S., Yoshinaga, N., Toyoda, M., Kitsuregawa, M.: Modeling situations in neural chat bots …

Observing dialogue in therapy: Categorizing and forecasting behavioral codes
J Cao, M Tanana, ZE Imel, E Poitras, DC Atkins… – arXiv preprint arXiv …, 2019 – arxiv.org
… therapy dialogue. 1 Introduction Conversational agents have long been studied in the context of psychotherapy, going back to chat- bots such as ELIZA (Weizenbaum, 1966) and PARRY (Colby, 1975). Research in modeling …

Natural Language Processing and Chinese Computing: 8th CCF International Conference, NLPCC 2019, Dunhuang, China, October 9–14, 2019 …
J Tang, MY Kan, D Zhao, S Li, H Zan – 2019 – books.google.com
Page 1. Jie Tang· Min-Yen Kan · Dongyan Zhao · Sujian Li· Hongying Zan (Eds.) Natural Language Processing and Chinese Computing 8th CCF International Conference, NLPCC 2019 Dunhuang, China, October 9–14, 2019 Proceedings, Part I 123 Page 2 …

Socially-Aware Dialogue System
R Zhao – 2019 – lti.cs.cmu.edu
… However, social chatbots fall short in replicating the interpersonal function of communication … SAPA) This chapter reviews our knowledge-inspired socially-aware dialogue system in a … recognition of conversational strategies in the service of a socially-aware dialog system …

A comparative study of word embedding methods for early risk prediction on the Internet
E Fano – 2019 – diva-portal.org
… Another realistic scenario would be to provide support for help lines and hospitals. It would be possible to develop chat bots and other dialogue systems that can determine the severity of a person’s mental health risk based on just a few lines of text …

Have a chat with BERT; passage re-ranking using conversational context
T Crijns, A de Vries – 2019 – ru.nl
… In technologically ad- vanced societies, it is common to encounter such artificial conversation partners, as the underlying technologies have existed in various forms for some time since as early as 1966, when the pioneering chatbot ELIZA was created in the MIT AI Labo- ratory …

Fine-tuning language models from human preferences
DM Ziegler, N Stiennon, J Wu, TB Brown… – arXiv preprint arXiv …, 2019 – arxiv.org
… learning from language. RL is not the only way to incorporate ongoing human feedback: Hancock et al. (2019) ask humans what a dialogue system should have said instead, then continue supervised training. In this paper, we …

Trouble on the horizon: Forecasting the derailment of online conversations as they develop
JP Chang, C Danescu-Niculescu-Mizil – arXiv preprint arXiv:1909.01362, 2019 – arxiv.org
… with these properties already exist, albeit geared toward generation rather than prediction: recent work in context-aware dialog generation (or “chat- bots”) has proposed … The former can be pre-trained on large amounts of unsuper- vised data, similarly to how chatbots are trained …

Efficient Algorithm for Answering Fact-based Queries Using Relational Data Enriched by Context-Based Embeddings
AA Altowayan – 2019 – webpage.pace.edu
… Page 3. Abstract Intelligent conversational systems – such as question answering and chatbots – are becoming a more critical component of today’s AI in areas ranging from health, medicine, and security, to personal assistants, and other domains …

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language …
K Inui, J Jiang, V Ng, X Wan – Proceedings of the 2019 Conference on …, 2019 – aclweb.org
Page 1. EMNLP-IJCNLP 2019 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing Proceedings of the Conference November 3–7, 2019 Hong Kong, China Page 2 …

Adaptive and Personalized Systems Based on Semantics
P Lops, C Musto, F Narducci, G Semeraro – Semantics in Adaptive and …, 2019 – Springer
In the introduction of this book, we have thoroughly discussed the importance of adaptive and personalized systems in a broad range of applications. In particular, we have motivated the use of…

Structured Knowledge Discovery from Massive Text Corpus
C Zhang – arXiv preprint arXiv:1908.01837, 2019 – arxiv.org
… and chat-bots become more and more popular, users may ask smart devices questions via … Figure 1 illustrates three scenarios on community Q&A, voice assistant/chatbot, and service … text classification tasks where the label of text is highly correlated with some topic-specific …

Adaptace jazykového modelu na téma v reálném ?ase
J Lehe?ka – 2019 – otik.uk.zcu.cz
… 59 6.5 Supervised Text Classification … days. Typical applications, in which ASR is indispensable, are: (1) chatbots, virtual assis- tants and dialogue systems, where people are conversing with computers, (2) dictating …

Deep learning for nlp and speech recognition
U Kamath, J Liu, J Whitaker – 2019 – Springer
… We introduce language modeling and discuss applications such as text classification, clustering, machine translation, question answering, automatic summarization, and automated speech recognition, con- cluding with a case study on text clustering and topic modeling …

ICTAI 2019
YM Boumarafi, Y Salhi – computer.org
… TECH (ERIS)),xxiii, , ,NLP / NLU – 3,,Sentiment-Aware Short Text Classification Based on … Inner Mongolia University),FPSeq: Simplifying and Accelerating Task-Oriented Dialogue Systems via Fully … of Argument Type and Concerns in Argumentation with a Chatbot , 1549, Lisa …

Automatically responding to customers
R Huijzer – pure.tue.nl
… The IBM sales department claims that Autodesk using chatbots cut down their resolution … Conversational agent or dialogue systems aim to communicate with humans using natural lan- guage. Consensus is not clear on whether a chatbot is synonymous to conversational agent …

Automating app review response generation
C Gao, J Zeng, X Xia, D Lo, MR Lyu… – 2019 34th IEEE/ACM …, 2019 – ieeexplore.ieee.org
… the focus of our work. Dialogue generation has been extensively studied in the natural language processing field [10]–[12], for facilitating social conversations, eg, the Microsoft XiaoIce chatbot [13]. Such work is generally grounded …

Machine Learning from Casual Conversation
A Mohammed Ali – 2019 – stars.library.ucf.edu
… 2.1 Conversational Agents and Chatbots There is a long history of studying conversational agents and chatbots. The earliest known chatbot was ELIZA [132], which was designed to emulate a Rogerian therapist. To provide its responses …

Multimodal integration for interactive conversational systems
M Johnston – The Handbook of Multimodal-Multisensor Interfaces …, 2019 – dl.acm.org
Page 1. IPART MULTIMODAL LANGUAGE AND DIALOGUE PROCESSING Page 2. Page 3. 1Multimodal Integration for Interactive Conversational Systems Michael Johnston 1.1 Introduction This chapter discusses the challenges …

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
A Korhonen, D Traum, L Màrquez – … of the 57th Annual Meeting of the …, 2019 – aclweb.org
Page 1. ACL 2019 The 57th Annual Meeting of the Association for Computational Linguistics Proceedings of the Conference July 28 – August 2, 2019 Florence, Italy Page 2. Diamond Sponsors: Platinum Sponsors: ii Page 3. Gold sponsors: Keiosk Analytics Silver sponsors …

Artificial Intelligence in the legal sector. A comparative analysis of expert and AI approaches to predicting court decisions
N Kaliazina – 2019 – pdfs.semanticscholar.org
… ones such as spell checking, keyword search, information extraction from websites, documents classification, machine translation, spoken dialogue systems, and complex … because for both cases the solution lies in the binary text classification. Therefore, it is … chatbots that …

Response Retrieval in Information-seeking Conversations
L Yang – 2019 – scholarworks.umass.edu
Page 1. University of Massachusetts Amherst ScholarWorks@UMass Amherst Doctoral Dissertations Dissertations and Theses 2019 Response Retrieval in Information-seeking Conversations Liu Yang College of Information and Computer Sciences, UMass Amherst …

Artificial Intelligence in Education
S Isotani, E Millán, A Ogan, P Hastings, B McLaren… – 2019 – Springer
Page 1. Seiji Isotani · Eva Millán · Amy Ogan · Peter Hastings · Bruce McLaren · Rose Luckin (Eds.) 123 LNAI 11626 20th International Conference, AIED 2019 Chicago, IL, USA, June 25-29, 2019 Proceedings, Part II Artificial Intelligence in Education Page 2 …

Contextual language understanding Thoughts on Machine Learning in Natural Language Processing
B Favre – 2019 – hal-amu.archives-ouvertes.fr
… General purpose dialog agents, also known as “chatbots”, are a good example of how … The ELIZA chatbot (Weizenbaum 1976) or contestants to the Loeb- ner Prize competition (Stephens … valency assessment, predicting the quantity of silence required for a dialog system to take …

ArgueBot: Enabling debates through a hybrid retrieval-generation-based chatbot
I Kulatska – 2019 – essay.utwente.nl
… 2.2 Chatbots 9 Page 15. like the chatbot is. Chatbots can also be evaluated by conducting user tests and using surveys to determine user satisfaction (Higashinaka et al., 2018). 2.3 Conclusion Concluding the literature review …

Learning to Converse With Latent Actions
T Zhao – 2019 – lti.cs.cmu.edu
… 2.1 Dialog system pipeline for task-oriented dialog systems . . . . . 8 … These unique features make latent action E2E dialog system powerful and practical for creating dialog systems in a variety of usage and domains. 1.2 Thesis Statement …

Neural architectures for natural language understanding
Y Tay – 2019 – dr.ntu.edu.sg
Page 1. Neural Architectures for Natural Language Understanding Tay Yi School of Computer Science & Engineering A thesis submitted to the Nanyang Technological University in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2019 Page 2 …

Computational Framework for Facilitating Intimate Dyadic Communication
D Utami – 2019 – repository.library.northeastern.edu
… In Chapter 4, I introduced the research platform used throughout the experiments in the dissertation, including the Furhat robot and the IrisTK dialog system. In Chapter 5, I answer the first research question by presenting a study on the acceptance …