Discourse & Dialogue
Topic Modeling -- Intro and Implementation
Businesses interact with their customers to better understand them and also to improve their products and services. This interaction can take the form of emails, textual social media posts (e.g. It would be inefficient and cost-prohibitive to have human representatives look through all of these forms of textual communications and then route the communications to the relevant teams to review, take action on and/or respond to customers. One inexpensive method to group such interactions and to assign them to relevant teams is using topic modeling. Topic modeling in the context of Natural Language Processing (NLP) is a type of unsupervised (i.e.
Analysis and Utilization of Entrainment on Acoustic and Emotion Features in User-agent Dialogue
Tan, Daxin, Kargas, Nikos, McHardy, David, Papayiannis, Constantinos, Bonafonte, Antonio, Strelec, Marek, Rohnke, Jonas, Filandras, Agis Oikonomou, Wood, Trevor
Entrainment is the phenomenon by which an interlocutor adapts their speaking style to align with their partner in conversations. It has been found in different dimensions as acoustic, prosodic, lexical or syntactic. In this work, we explore and utilize the entrainment phenomenon to improve spoken dialogue systems for voice assistants. We first examine the existence of the entrainment phenomenon in human-to-human dialogues in respect to acoustic feature and then extend the analysis to emotion features. The analysis results show strong evidence of entrainment in terms of both acoustic and emotion features. Based on this findings, we implement two entrainment policies and assess if the integration of entrainment principle into a Text-to-Speech (TTS) system improves the synthesis performance and the user experience. It is found that the integration of the entrainment principle into a TTS system brings performance improvement when considering acoustic features, while no obvious improvement is observed when considering emotion features.
Why Companies Should Invest in Sentiment Analysis
Monitoring and examining sentiments have become increasingly popular with brands focused on automating their business processes. Mainly known as an innovative tool used by social media and marketing analysts, sentiment analysis, sometimes referred to as "social listening," has also proved helpful in other functional areas. We explain why companies should invest in sentiment analysis. Insight engines allow to use sentiment analysis across the enterprise and doesn't limit the tool to just one business need. Without machine learning (ML), methods like natural language processing (NLP) sentiment analysis would be unachievable.
A Transformer-Based User Satisfaction Prediction for Proactive Interaction Mechanism in DuerOS
Shen, Wei, He, Xiaonan, Zhang, Chuheng, Zhang, Xuyun, XIe, Jian
Recently, spoken dialogue systems have been widely deployed in a variety of applications, serving a huge number of end-users. A common issue is that the errors resulting from noisy utterances, semantic misunderstandings, or lack of knowledge make it hard for a real system to respond properly, possibly leading to an unsatisfactory user experience. To avoid such a case, we consider a proactive interaction mechanism where the system predicts the user satisfaction with the candidate response before giving it to the user. If the user is not likely to be satisfied according to the prediction, the system will ask the user a suitable question to determine the real intent of the user instead of providing the response directly. With such an interaction with the user, the system can give a better response to the user. Previous models that predict the user satisfaction are not applicable to DuerOS which is a large-scale commercial dialogue system. They are based on hand-crafted features and thus can hardly learn the complex patterns lying behind millions of conversations and temporal dependency in multiple turns of the conversation. Moreover, they are trained and evaluated on the benchmark datasets with adequate labels, which are expensive to obtain in a commercial dialogue system. To face these challenges, we propose a pipeline to predict the user satisfaction to help DuerOS decide whether to ask for clarification in each turn. Specifically, we propose to first generate a large number of weak labels and then train a transformer-based model to predict the user satisfaction with these weak labels. Empirically, we deploy and evaluate our model on DuerOS, and observe a 19% relative improvement on the accuracy of user satisfaction prediction and 2.3% relative improvement on user experience.
Video Games as a Corpus: Sentiment Analysis using Fallout New Vegas Dialog
Hämäläinen, Mika, Alnajjar, Khalid, Poibeau, Thierry
We present a method for extracting a multilingual sentiment annotated dialog data set from Fallout New Vegas. The game developers have preannotated every line of dialog in the game in one of the 8 different sentiments: \textit{anger, disgust, fear, happy, neutral, pained, sad } and \textit{surprised}. The game has been translated into English, Spanish, German, French and Italian. We conduct experiments on multilingual, multilabel sentiment analysis on the extracted data set using multilingual BERT, XLMRoBERTa and language specific BERT models. In our experiments, multilingual BERT outperformed XLMRoBERTa for most of the languages, also language specific models were slightly better than multilingual BERT for most of the languages. The best overall accuracy was 54\% and it was achieved by using multilingual BERT on Spanish data. The extracted data set presents a challenging task for sentiment analysis. We have released the data, including the testing and training splits, openly on Zenodo. The data set has been shuffled for copyright reasons.
Grammar Detection for Sentiment Analysis through Improved Viterbi Algorithm
Chavali, Surya Teja, Kandavalli, Charan Tej, M, Sugash T
Grammar Detection, also referred to as Parts of Speech Tagging of raw text, is considered an underlying building block of the various Natural Language Processing pipelines like named entity recognition, question answering, and sentiment analysis. In short, forgiven a sentence, Parts of Speech tagging is the task of specifying and tagging each word of a sentence with nouns, verbs, adjectives, adverbs, and more. Sentiment Analysis may well be a procedure accustomed to determining if a given sentence's emotional tone is neutral, positive or negative. To assign polarity scores to the thesis or entities within phrase, in-text analysis and analytics, machine learning and natural language processing, approaches are incorporated. This Sentiment Analysis using POS tagger helps us urge a summary of the broader public over a specific topic. For this, we are using the Viterbi algorithm, Hidden Markov Model, Constraint based Viterbi algorithm for POS tagging. By comparing the accuracies, we select the foremost accurate result of the model for Sentiment Analysis for determining the character of the sentence.
KPT: Keyword-guided Pre-training for Grounded Dialog Generation
Zhu, Qi, Mi, Fei, Zhang, Zheng, Wang, Yasheng, Li, Yitong, Jiang, Xin, Liu, Qun, Zhu, Xiaoyan, Huang, Minlie
Incorporating external knowledge into the response generation process is essential to building more helpful and reliable dialog agents. However, collecting knowledge-grounded conversations is often costly, calling for a better pre-trained model for grounded dialog generation that generalizes well w.r.t. different types of knowledge. In this work, we propose KPT (Keyword-guided Pre-Training), a novel self-supervised pre-training method for grounded dialog generation without relying on extra knowledge annotation. Specifically, we use a pre-trained language model to extract the most uncertain tokens in the dialog as keywords. With these keywords, we construct two kinds of knowledge and pre-train a knowledge-grounded response generation model, aiming at handling two different scenarios: (1) the knowledge should be faithfully grounded; (2) it can be selectively used. For the former, the grounding knowledge consists of keywords extracted from the response. For the latter, the grounding knowledge is additionally augmented with keywords extracted from other utterances in the same dialog. Since the knowledge is extracted from the dialog itself, KPT can be easily performed on a large volume and variety of dialogue data. We considered three data sources (open-domain, task-oriented, conversational QA) with a total of 2.5M dialogues. We conduct extensive experiments on various few-shot knowledge-grounded generation tasks, including grounding on dialog acts, knowledge graphs, persona descriptions, and Wikipedia passages. Our comprehensive experiments and analyses demonstrate that KPT consistently outperforms state-of-the-art methods on these tasks with diverse grounding knowledge.
Discourse and conversation impairments in patients with dementia
Dementia is the progressive deterioration of cognitive, linguistic, and social functioning that affects the quality of life, including the physical, social, and economic conditions of individuals, their families, and society (2-5). Although there is no treatment for dementia, early-stage identification, and assessment of individuals with dementia are of utmost importance to enable interventions that can delay the progression of dementia and support family planning. The neurocognitive assessment aims to evaluate individuals' condition and provide early diagnosis, prognosis, and quantify intervention efficacy. Speech, language, and communication impairments are early symptoms in individuals with dementia (6-8). For example, earlier studies have shown that discourse narratives in the autobiographies of Catholic sisters of the School Sisters of Notre Dame congregation can be an exceedingly early predictor of dementia (9). In addition, studies of the speeches of the US president Ronald Reagan (10, 11) and the comparative analysis of the British novelists Iris Murdoch and Agatha Christie works showed that narratives could provide an early prognosis of dementia development (12). Clinical Discourse Analysis (CDA) examines speech, language, and communication impairments in individuals with dementia and elicits language and communication measures. These measures can provide an early, stressless, and comprehensive assessment of individuals' language and neurocognitive functioning (e.g., memory, attention, social interaction) and inform treatment approaches (13, 14). CDA involves the characterization of texts produced by individuals through language, cooperation, and social interaction in communicative settings such as conversations, semi-structured interviews (15-17), role-plays, and monologues (18).
Improving Sentiment Analysis By Emotion Lexicon Approach on Vietnamese Texts
The sentiment analysis task has various applications in practice. In the sentiment analysis task, words and phrases that represent positive and negative emotions are important. Finding out the words that represent the emotion from the text can improve the performance of the classification models for the sentiment analysis task. In this paper, we propose a methodology that combines the emotion lexicon with the classification model to enhance the accuracy of the models. Our experimental results show that the emotion lexicon combined with the classification model improves the performance of models.
Twitter Data Analysis: Izmir Earthquake Case
Agrali, Özgür, Sökün, Hakan, Karaarslan, Enis
T\"urkiye is located on a fault line; earthquakes often occur on a large and small scale. There is a need for effective solutions for gathering current information during disasters. We can use social media to get insight into public opinion. This insight can be used in public relations and disaster management. In this study, Twitter posts on Izmir Earthquake that took place on October 2020 are analyzed. We question if this analysis can be used to make social inferences on time. Data mining and natural language processing (NLP) methods are used for this analysis. NLP is used for sentiment analysis and topic modelling. The latent Dirichlet Allocation (LDA) algorithm is used for topic modelling. We used the Bidirectional Encoder Representations from Transformers (BERT) model working with Transformers architecture for sentiment analysis. It is shown that the users shared their goodwill wishes and aimed to contribute to the initiated aid activities after the earthquake. The users desired to make their voices heard by competent institutions and organizations. The proposed methods work effectively. Future studies are also discussed.