Predicting Similarities in User Profiles and Making Search Predictions
Can we look at individual search sessions and identify users who are similar in terms of search goals and behaviors? One of the potential goals of this research would be to identify users who are similar and predict their search outcomes (success or failure). Another goal is to make suggestions on what information could be potentially useful for the user (information fostering).
Improving Dialogue in User-System Interactions
Information search is no longer limited to typing queries in the search box. Query-less searches, common in voice-based assistants, make use of natural language conversations between the user and the system. With increased knowledge about the capabilities of the system and the requirements of the user, the system understands how to improve the search results. This is a result of increased personalization and better relevance assessments. This project will aim to improve the dialogue or conversations which occur between the user and the system during an information gathering session.
Question Reformulations in Community Q&A
Automatic reformulation of questions in Community Question Answering (CQA) sites in an effort to improve the quality of questions. It is common for CQA sites to discard poor-quality questions. The goal of this research is to design an approach to automatically reformulate such questions, with or without any feedback from the users, to match the quality requirements of the site. The system is to be trained using questions which are similar and have been frequently upvoted and answered in the community.
Generating Information Behaviors
Collecting and identifying micro-actions that comprise information seeking sessions, and then recombining them in order to create novel sessions and potentially improved outcomes. These actions comprise larger information behaviors such as browsing, but can be arranged into different patterns thus creating unique and previously unidentified behaviors. This project is based on the research into digital infinity and the possibility for a small set of ordered items to create near infinite possibilities.
Summarizing Collections
Generating summaries in multiple formats including tables, narratives, abstracts, and other visualizations from a collection of documents, with a particular focus on government and financial documents. The goal is to take advantage of underlying textual structures, such as formatting and tagging, as well as implied structures, such as time series, hierarchical relationships and other dependencies.
Answer generation
Generating automated multimedia answers in the form of text, appropriate image, videos from the Community Question Answering (CQA) sites, with a focus on answer quality. The goal is to design an algorithm which will take account of the question type and its intended audience to generate automated answers. The source of the answers can be some combination of existing answers provided to previously asked similar questions, or some other resources depending on the context.
Our Amazing Team
Chirag Shah
Group Director, iSchool and Computer Science faculty
Information seeking; Social media; Data analytics
Souvick Ghosh
Group Leader, iSchool Ph.D. Student
Conversational IR, Deep Learning, NLP, Text Analysis
Jonathan Pulliza
Text Mining; Machine Learning; NLP
Soumik Mandal
Machine Learning; NLP; Human Information Behavior
Ruoyuan Gao
Computer Science PhD Student
Sophia Song
Undergraduate Student
Nithi Kumar
Undergraduate Student
Arvind Yalavarti
Past member
About Us
NeuIR (pronounced “New IR”) Group is focused on solving problems at the intersection of Deep Neural Network, Big Data, and Multimedia Information Retrieval. We understand the importance of different forms of data (text, speech, images, sensor data), and strive to achieve the following research goals:
(a) Exploring different deep learning techniques in an effort to advance artificial intelligence in information retrieval;
(b) Identifying challenges in handling big data and developing best practices to solve these problems;
(c) Developing newer retrieval models and evaluation metrics specific to conversational information retrieval.
Ghosh, S., & Shah, C. (2018). Towards Automatic Fake News Classification. Proceedings of the Association for Information Science and Technology (accepted).
Mandal, S., Rath, M., Wang, Y., & Patra, B. G. (2018, March). Predicting Zika Prevention Techniques Discussed on Twitter: An Exploratory Study. InProceedings of the 2018 Conference on Human Information Interaction&Retrieval (pp. 269-272). ACM.
Shah, C. (2018, March). Information Fostering-Being Proactive with Information Seeking and Retrieval: Perspective Paper. In Proceedings of the 2018 Conference on Human Information Interaction&Retrieval (pp. 62-71). ACM.
Singh, V. K., Ghosh, S., & Jose, C. (2017, May). Toward multimodal cyberbullying detection. In Proceedings of the 2017 CHI Conference Extended Abstracts on Human Factors in Computing Systems (pp. 2090-2099). ACM.
With more than 5 billion active phone subscriptions, and related trends in objective-self, smart health, and sensor-based analysis, human lives are being captured at scale and resolution never possible before. This data allows us to study multiple human phenomena and understand human behavior in ways not possible before. An ability to understand human behavior including activity levels, location patterns, sleep, consumption, and communication and social interaction allows for highly detailed and personalized data collection in a far more granular, unobtrusive, and affordable way than earlier.
This research project is investigating the interconnections between human social and mobility behaviors and general wellbeing, self-esteem, and mental health. Specifically, the project is investigating how socio-mobile data (such as GPS locations and call logs) from users’ personal mobile devices can be used to understand the interplay between socio-mobile behavior and wellbeing, and how any changes in user wellbeing can be automatically detected via this data, which may potentially lead to future appropriate interventions.
In this study, detailed socio-mobile logs for about 60 participants were collected for 10 weeks on Rutgers campus during Spring 2015. The same participants were asked to fill in a number of surveys related to wellbeing. This project has resulted in multiple publications on the aspects of predicting cooperation and privacy needs. Data on other aspects (e.g. mental health) are being analyzed and will be released publically soon.
Understanding and Shaping Privacy Behavior
People have a fundamental right to maintain their personal privacy while using all kinds of technologies in their daily lives. Violations of this right can have extremely negative consequences for an individual, including trauma, depression, loss of dignity, loss of jobs, and in extreme cases loss of life. While people generally have some awareness that their personal data are being collected through their online social networks, mobile devices, and IoT sensors, they are often unaware of the extent and nature of the data collected through these technologies and express surprise and discomfort when they find out. For example, multiple users recently reported surprise when they found out that their flashlight app was continuously tracking their location information. Our group has been studying how the newer technologies are creating newer privacy challenges as well as opportunities for mitigation. Some recent projects include:
Uniqueness of personal spending data
Identifying the limits of anonymity in the emerging big data. For example, in a recent Science paper, we reported that it takes only 4 pieces of spatio-temporal information to uniquely identify a customer’s credit card data amongst millions of other anonymized data records. This brings to the fore an urgent need to re-define how we measure privacy, understand it, and ultimately what drives our privacy needs in an information rich eco-system.
Predicting privacy attitudes using phone metadata
We recently found that a combination of phone use metadata can provide vital insights into a person’s privacy attitudes. This can be used to automatically recommend privacy settings to individuals.
Nudging privacy behavior
We took inspiration from “second-hand” smoking cessation campaigns to create interfaces that made users realize that each time they leak information about themselves, by proxy, they also leak information about their peers. We found that individuals thought longer and were twice less likely to give away personal information when made aware of such social consequences of their actions.
Understanding and Preventing Cyberbullying
We are working on two projects related to cyberbullying prevention.
Focus Group Studies to Understand Cyberbullying in Schools.
This exploratory work studies the effects of emerging app features on the cyberbullying practices in high school settings. These include the increasing prevalence of image/video content, perceived ephemerality, anonymity, and hyperlocal communication. Based on qualitative analysis of focus groups and follow-up individual interviews with high school students, these features were found to influence the practice of cyberbullying, as well as creating negative socio-psychological effects. For example, visual data was found to be used in cyberbullying settings as evidence of contentious events, a repeated reminder, and caused a graphic impact on recipients. Similarly, perceived ephemerality of content was found to be associated with “broken expectations” with respect to the apps and severe bullying outcomes for those affected. Results shed light on an important technology-mediated social phenomenon of cyberbullying, improve understanding of app use (and abuse) by the teenage user population, and pave the way for future research on countering app-centric cyberbullying.
Cyberbullying Detection Using Text and Social Network Analysis
This project aims to define new approaches for automatic detection of cyberbullying by integrating the relevant research in social sciences and computer science. Cyberbullying is a critical social problem that occurs over a technical substrate. According to a recent National Crime Prevention Council report, more than 40% of teenagers in the US have reported being cyberbullied. This is especially worrying as the multiple studies have reported that the victims of cyberbullying often deal with psychiatric and psychosomatic disorders. Specifically, this research will advance the state of the art in cyberbullying detection beyond textual analysis by also giving due attention to the social relationships in which these bullying messages are exchanged. A higher accuracy at detection would allow for better mitigation of the cyberbullying phenomenon and may help improve the lives of thousands of victims who are cyberbullied each year. The results of this research will also open doors to employing social intervention mechanisms to help prevent cyberbullying incidents in future. The findings from this research will also validate and refine existing theories on cyberbullying and potentially advance the field by creating a wave of data-driven analysis of the phenomenon. The generated data set will be made available to the larger research community, thus enabling new findings that can help counter this social problem.
Qianjia Huang, Vivek Kumar Singh, and Pradeep Kumar Atrey. Cyberbullying detection using social network analysis. Submitted to: IEEE Transactions on Computational Social
We are happy to share the cyberbullying labeled dataset with other interested researchers. We would ask you to sign an agreement respecting the privacy of the users in the dataset. Please email Prof. Vivek Singh (v.singh@rutgers.edu) to request the dataset.
