Discourse & Dialogue
T-BERT -- Model for Sentiment Analysis of Micro-blogs Integrating Topic Model and BERT
Palani, Sarojadevi, Rajagopal, Prabhu, Pancholi, Sidharth
Sentiment analysis (SA) has become an extensive research area in recent years impacting diverse fields including ecommerce, consumer business, and politics, driven by increasing adoption and usage of social media platforms. It is challenging to extract topics and sentiments from unsupervised short texts emerging in such contexts, as they may contain figurative words, strident data, and co-existence of many possible meanings for a single word or phrase, all contributing to obtaining incorrect topics. Most prior research is based on a specific theme/rhetoric/focused-content on a clean dataset. In the work reported here, the effectiveness of BERT(Bidirectional Encoder Representations from Transformers) in sentiment classification tasks from a raw live dataset taken from a popular microblogging platform is demonstrated. A novel T-BERT framework is proposed to show the enhanced performance obtainable by combining latent topics with contextual BERT embeddings. Numerical experiments were conducted on an ensemble with about 42000 datasets using NimbleBox.ai platform with a hardware configuration consisting of Nvidia Tesla K80(CUDA), 4 core CPU, 15GB RAM running on an isolated Google Cloud Platform instance. The empirical results show that the model improves in performance while adding topics to BERT and an accuracy rate of 90.81% on sentiment classification using BERT with the proposed approach.
How to Create and Deploy a Simple Sentiment Analysis App via API - KDnuggets
Let's say you've built an NLP model for some specific task, whether it be text classification, question answering, translation, or what have you. You've tested it out locally and it performs well. You've had others test it out as well, and it continues to perform well. Now you want to roll it out to a larger audience, be that audience a team of developers you work with, a specific group of end users, or even the general public. You have decided that you want to do so using a REST API, as you find this to be your best option.
DialoGraph: Incorporating Interpretable Strategy-Graph Networks into Negotiation Dialogues
Joshi, Rishabh, Balachandran, Vidhisha, Vashishth, Shikhar, Black, Alan, Tsvetkov, Yulia
To successfully negotiate a deal, it is not enough to communicate fluently: pragmatic planning of persuasive negotiation strategies is essential. While modern dialogue agents excel at generating fluent sentences, they still lack pragmatic grounding and cannot reason strategically. We present DialoGraph, a negotiation system that incorporates pragmatic strategies in a negotiation dialogue using graph neural networks. DialoGraph explicitly incorporates dependencies between sequences of strategies to enable improved and interpretable prediction of next optimal strategies, given the dialogue context. Our graph-based method outperforms prior state-of-the-art negotiation models both in the accuracy of strategy/dialogue act prediction and in the quality of downstream dialogue response generation. We qualitatively show further benefits of learned strategy-graphs in providing explicit associations between effective negotiation strategies over the course of the dialogue, leading to interpretable and strategic dialogues.
Validating GAN-BioBERT: A Methodology For Assessing Reporting Trends In Clinical Trials
Myszewski, Joshua J, Klossowski, Emily, Meyer, Patrick, Bevil, Kristin, Klesius, Lisa, Schroeder, Kristopher M
In the past decade, there has been much discussion about the issue of biased reporting in clinical research. Despite this attention, there have been limited tools developed for the systematic assessment of qualitative statements made in clinical research, with most studies assessing qualitative statements relying on the use of manual expert raters, which limits their size. Also, previous attempts to develop larger scale tools, such as those using natural language processing, were limited by both their accuracy and the number of categories used for the classification of their findings. With these limitations in mind, this study's goal was to develop a classification algorithm that was both suitably accurate and finely grained to be applied on a large scale for assessing the qualitative sentiment expressed in clinical trial abstracts. Additionally, this study seeks to compare the performance of the proposed algorithm, GAN-BioBERT, to previous studies as well as to expert manual rating of clinical trial abstracts. This study develops a three-class sentiment classification algorithm for clinical trial abstracts using a semi-supervised natural language process model based on the Bidirectional Encoder Representation from Transformers (BERT) model, from a series of clinical trial abstracts annotated by a group of experts in academic medicine. Results: The use of this algorithm was found to have a classification accuracy of 91.3%, with a macro F1-Score of 0.92, which is a significant improvement in accuracy when compared to previous methods and expert ratings, while also making the sentiment classification finer grained than previous studies. The proposed algorithm, GAN-BioBERT, is a suitable classification model for the large-scale assessment of qualitative statements in clinical trial literature, providing an accurate, reproducible tool for the large-scale study of clinical publication trends.
Sentiment analysis in tweets: an assessment study from classical to modern text representation models
Barreto, Sérgio, Moura, Ricardo, Carvalho, Jonnathan, Paes, Aline, Plastino, Alexandre
With the growth of social medias, such as Twitter, plenty of user-generated data emerge daily. The short texts published on Twitter -- the tweets -- have earned significant attention as a rich source of information to guide many decision-making processes. However, their inherent characteristics, such as the informal, and noisy linguistic style, remain challenging to many natural language processing (NLP) tasks, including sentiment analysis. Sentiment classification is tackled mainly by machine learning-based classifiers. The literature has adopted word representations from distinct natures to transform tweets to vector-based inputs to feed sentiment classifiers. The representations come from simple count-based methods, such as bag-of-words, to more sophisticated ones, such as BERTweet, built upon the trendy BERT architecture. Nevertheless, most studies mainly focus on evaluating those models using only a small number of datasets. Despite the progress made in recent years in language modelling, there is still a gap regarding a robust evaluation of induced embeddings applied to sentiment analysis on tweets. Furthermore, while fine-tuning the model from downstream tasks is prominent nowadays, less attention has been given to adjustments based on the specific linguistic style of the data. In this context, this study fulfils an assessment of existing language models in distinguishing the sentiment expressed in tweets by using a rich collection of 22 datasets from distinct domains and five classification algorithms. The evaluation includes static and contextualized representations. Contexts are assembled from Transformer-based autoencoder models that are also fine-tuned based on the masked language model task, using a plethora of strategies.
Introduction to NLP with Disaster Tweets
Natural Language Processing, also known as NLP, is a subfield of computer science, specifically artificial intelligence, that focuses on understanding written and spoken text. It covers various tasks some of which are speech recognition, sentiment analysis and language generation; And, it has been applied in several use cases such as machine translation, spam detection, virtual assistants and chatbots. The project covered in this article is a sentiment analysis project called Natural Language Processing with Disaster Tweets. Sentiment analysis is the process to extract subjective qualities from text such as emotion or attitude. The objective of the project is to identify if a specific tweet is a real disaster or not. The project is ideal for beginners in NLP.
How companies use sentiment analysis to both ensure strong brand management
"Sentiment" is a rather intriguing concept. It can mean attitude, feeling, bias, view, thought, and even something as deeply felt as emotion. Sentiment analysis that utilizes AI and machine learning has become a powerful tool for companies to understand how their customers and/or potential customers feel about their company. It can also be used for competitive analysis to take the temperature of a rival's products or services. "When you can't convince them with intellect, persuade them with sentiment," is author Amit Kalantri's recommendation.
Non-negative matrix factorization algorithms greatly improve topic model fits
Carbonetto, Peter, Sarkar, Abhishek, Wang, Zihao, Stephens, Matthew
We report on the potential for using algorithms for non-negative matrix factorization (NMF) to improve parameter estimation in topic models. While several papers have studied connections between NMF and topic models, none have suggested leveraging these connections to develop new algorithms for fitting topic models. Importantly, NMF avoids the "sum-to-one" constraints on the topic model parameters, resulting in an optimization problem with simpler structure and more efficient computations. Building on recent advances in optimization algorithms for NMF, we show that first solving the NMF problem then recovering the topic model fit can produce remarkably better fits, and in less time, than standard algorithms for topic models. While we focus primarily on maximum likelihood estimation, we show that this approach also has the potential to improve variational inference for topic models. Our methods are implemented in the R package fastTopics.
Multi-turn Dialog System on Single-turn Data in Medical Domain
Sorathiya, Nazib, Lin, Chuan-An, Xiong, Daniel Chen Daniel, Zin, Scott, Zhang, Yi, Yang, He Sarina, Huang, Sharon Xiaolei
Recently there has been a huge interest in dialog systems. This interest has also been developed in the field of the medical domain where researchers are focusing on building a dialog system in the medical domain. This research is focused on the multi-turn dialog system trained on the multi-turn dialog data. It is difficult to gather a huge amount of multi-turn conversational data in the medical domain that is verified by professionals and can be trusted. However, there are several frequently asked questions (FAQs) or single-turn QA pairs that have information that is verified by the experts and can be used to build a multi-turn dialog system.
VISITRON: Visual Semantics-Aligned Interactively Trained Object-Navigator
Shrivastava, Ayush, Gopalakrishnan, Karthik, Liu, Yang, Piramuthu, Robinson, Tür, Gokhan, Parikh, Devi, Hakkani-Tür, Dilek
Interactive robots navigating photo-realistic environments face challenges underlying vision-and-language navigation (VLN), but in addition, they need to be trained to handle the dynamic nature of dialogue. However, research in Cooperative Vision-and-Dialog Navigation (CVDN), where a navigator interacts with a guide in natural language in order to reach a goal, treats the dialogue history as a VLN-style static instruction. In this paper, we present VISITRON, a navigator better suited to the interactive regime inherent to CVDN by being trained to: i) identify and associate object-level concepts and semantics between the environment and dialogue history, ii) identify when to interact vs. navigate via imitation learning of a binary classification head. We perform extensive ablations with VISITRON to gain empirical insights and improve performance on CVDN. VISITRON is competitive with models on the static CVDN leaderboard. We also propose a generalized interactive regime to fine-tune and evaluate VISITRON and future such models with pre-trained guides for adaptability.