TweetGrep: Weakly Supervised Joint Retrieval and Sentiment Analysis of Topical Tweets

AAAI Conferences

An overwhelming amount of data is generated everyday onsocial media, encompassing a wide spectrum of topics. With almost every business decision depending on customer opinion, mining of social media data needs to be quick and easy.For a data analyst to keep up with the agility and the scale of the data, it is impossible to bank on fully supervised techniques to mine topics and their associated sentiments from social media. Motivated by this, we propose a weakly supervised approach (named, TweetGrep) that lets the data analyst easily define a topic by few keywords and adapt a generic sentiment classifier to the topic – by jointly modeling topics and sentiments using label regularization. Experiments with diverse datasets show that TweetGrep beats the state-of-the-art models for both the tasks of retrieving topical tweet sand analyzing the sentiment of the tweets (average improvement of 4.97% and 6.91% respectively in terms of area under the curve). Further, we show that TweetGrep can also be adopted in a novel task of hashtag disambiguation, which significantly outperforms the baseline methods.


Managing Bad News in Social Media: A Case Study on Domino’s Pizza Crisis

AAAI Conferences

Social media has become prominently popular. Tens of millions of users login to social media sites like Twitter to disseminate breaking news and share their opinions and thoughts. For businesses, social media is potentially useful for monitoring the public perception and the social reputation of companies and products. Despite great potential, how bad news about a company influences the public sentiments in social media has not been studied in depth. The aim of this study is to assess people’s sentiments in Twitter upon the spread of two types of information: corporate bad news and a CEO’s apology. We attempted to understand how sentiments on corporate bad news propagate in Twitter and whether any social network feature facilitates its spread. We investigated the Domino’s Pizza crisis in 2009, where bad news spread rapidly through social media followed by an official apology from the company. Our work shows that bad news spreads faster than other types of information, such as an apology, and sparks a great degree of negative sentiments in the network. However, when users converse about bad news repeatedly, their negative sentiments are softened. We discuss various reactions of users towards the bad news in social media such as negative purchase intent.


“Dancing with the Stars,” NBA Games, Politics: An Exploration of Twitter Users’ Response to Events

AAAI Conferences

Microblogging services such as Twitter offer great opportunities for analyzing the reactions of a wide audience with respect to current events. In this paper, we explore the correlation between types of user engagement and events centered around celebrities (e.g., personal or professional events involving Actors, Musicians, Politicians, Athletes).


Mossberg: The Disappearing Computer

#artificialintelligence

The biggest hardware and software arrival since the iPad in 2010 has been Amazon's Echo voice-controlled intelligent speaker, powered by its Alexa software assistant. But just because you're not seeing amazing new consumer tech products on Amazon, in the app stores, or at the Apple Store or Best Buy, that doesn't mean the tech revolution is stuck or stopped. They are: Artificial intelligence / machine learning, augmented reality, virtual reality, robotics and drones, smart homes, self-driving cars, and digital health / wearables. Google has changed its entire corporate mission to be "AI first" and, with Google Home and Google Assistant, to perform tasks via voice commands and eventually hold real, unstructured conversations.


SentiCite: An Approach for Publication Sentiment Analysis

arXiv.org Machine Learning

Abstract: With the rapid growth in the number of scientific publications, year after year, it is becoming increasingly difficult to identify quality authoritative work on a single topic. Though there is an availability of scientometric measures which promise to offer a solution to this problem, these measures are mostly quantitative and rely, for instance, only on the number of times an article is cited. With this approach, it becomes irrelevant if an article is cited 10 times in a positive, negative or neutral way. In this context, it is quite important to study the qualitative aspect of a citation to understand its significance. This paper presents a novel system for sentiment analysis of citations in scientific documents (SentiCite) and is also capable of detecting nature of citations by targeting the motivation behind a citation, e.g., reference to a dataset, reading reference. Furthermore, the paper also presents two datasets (SentiCiteDB and IntentCiteDB) containing about 2,600 citations with their ground truth for sentiment and nature of citation. SentiCite along with other state-of-the-art methods for sentiment analysis are evaluated on the presented datasets. Evaluation results reveal that SentiCite outperforms state-of-the-art methods for sentiment analysis in scientific publications by achieving a F1-measure of 0.71. 1 INTRODUCTION Sentiment analysis is the process of computationally categorizing and identifying opinions present in a textual document or images. As a field, sentiment analysis has been gaining a lot of interest from the scientific community in recent years. The main motivation for this work comes from the author's observation that there is an unavailability of a system capable of automatically analyzing the sentiment present in citations of scientific publications.