arXiv subjects

Shimei Pan

Publications and source records attributed to Shimei Pan.

30 records · Page 2Linked to original sources

arXiv2020

Causal Feature Selection with Dimension Reduction for Interpretable Text Classification

Text features that are correlated with class labels, but do not directly cause them, are sometimesuseful for prediction, but they may not be insightful. As an alternative to traditional correlation-basedfeature selection, causal inference could reveal more principled, meaningful relationships betweentext features and labels. To help researchers gain insight into text data, e.g. for social scienceapplications, in this paper we investigate a class of matching-based causal inference methods fortext feature selection. Features used in document classification are often high dimensional, howeverexisting causal feature selection methods use Propensity Score Matching (PSM) which is known to beless effective in high-dimensional spaces. We propose a new causal feature selection framework thatcombines dimension reduction with causal inference to improve text feature selection. Experiments onboth synthetic and real-world data demonstrate the promise of our methods in improving classificationand enhancing interpretability.

cs.LG↗

arXiv2020

Neural Fair Collaborative Filtering

A growing proportion of human interactions are digitized on social media platforms and subjected to algorithmic decision-making, and it has become increasingly important to ensure fair treatment from these algorithms. In this work, we investigate gender bias in collaborative-filtering recommender systems trained on social media data. We develop neural fair collaborative filtering (NFCF), a practical framework for mitigating gender bias in recommending sensitive items (e.g. jobs, academic concentrations, or courses of study) using a pre-training and fine-tuning approach to neural collaborative filtering, augmented with bias correction techniques. We show the utility of our methods for gender de-biased career and college major recommendations on the MovieLens dataset and a Facebook dataset, respectively, and achieve better performance and fairer behavior than several state-of-the-art models.

cs.IR↗

arXiv2019

An Intersectional Definition of Fairness

We propose definitions of fairness in machine learning and artificial intelligence systems that are informed by the framework of intersectionality, a critical lens arising from the Humanities literature which analyzes how interlocking systems of power and oppression affect individuals along overlapping dimensions including gender, race, sexual orientation, class, and disability. We show that our criteria behave sensibly for any subset of the set of protected attributes, and we prove economic, privacy, and generalization guarantees. We provide a learning algorithm which respects our intersectional fairness criteria. Case studies on census data and the COMPAS criminal recidivism dataset demonstrate the utility of our methods.

cs.LG↗

arXiv2019

Bayesian Modeling of Intersectional Fairness: The Variance of Bias

Intersectionality is a framework that analyzes how interlocking systems of power and oppression affect individuals along overlapping dimensions including race, gender, sexual orientation, class, and disability. Intersectionality theory therefore implies it is important that fairness in artificial intelligence systems be protected with regard to multi-dimensional protected attributes. However, the measurement of fairness becomes statistically challenging in the multi-dimensional setting due to data sparsity, which increases rapidly in the number of dimensions, and in the values per dimension. We present a Bayesian probabilistic modeling approach for the reliable, data-efficient estimation of fairness with multi-dimensional protected attributes, which we apply to two existing intersectional fairness metrics. Experimental results on census data and the COMPAS criminal justice recidivism dataset demonstrate the utility of our methodology, and show that Bayesian methods are valuable for the modeling and measurement of fairness in an intersectional context.

cs.LG↗

arXiv2019

Social Media-based User Embedding: A Literature Review

Automated representation learning is behind many recent success stories in machine learning. It is often used to transfer knowledge learned from a large dataset (e.g., raw text) to tasks for which only a small number of training examples are available. In this paper, we review recent advance in learning to represent social media users in low-dimensional embeddings. The technology is critical for creating high performance social media-based human traits and behavior models since the ground truth for assessing latent human traits and behavior is often expensive to acquire at a large scale. In this survey, we review typical methods for learning a unified user embeddings from heterogeneous user data (e.g., combines social media texts with images to learn a unified user representation). Finally we point out some current issues and future directions.

cs.SI↗

arXiv2018

Automatically Infer Human Traits and Behavior from Social Media Data

Given the complexity of human minds and their behavioral flexibility, it requires sophisticated data analysis to sift through a large amount of human behavioral evidence to model human minds and to predict human behavior. People currently spend a significant amount of time on social media such as Twitter and Facebook. Thus many aspects of their lives and behaviors have been digitally captured and continuously archived on these platforms. This makes social media a great source of large, rich and diverse human behavioral evidence. In this paper, we survey the recent work on applying machine learning to infer human traits and behavior from social media data. We will also point out several future research directions.

cs.SI↗

arXiv2017

Social Media-based Substance Use Prediction

In this paper, we demonstrate how the state-of-the-art machine learning and text mining techniques can be used to build effective social media-based substance use detection systems. Since a substance use ground truth is difficult to obtain on a large scale, to maximize system performance, we explore different feature learning methods to take advantage of a large amount of unsupervised social media data. We also demonstrate the benefit of using multi-view unsupervised feature learning to combine heterogeneous user information such as Facebook `"likes" and "status updates" to enhance system performance. Based on our evaluation, our best models achieved 86% AUC for predicting tobacco use, 81% for alcohol use and 84% for drug use, all of which significantly outperformed existing methods. Our investigation has also uncovered interesting relations between a user's social media behavior (e.g., word usage) and substance use.

cs.CL↗

arXiv2017

\$1 Today or \$2 Tomorrow? The Answer is in Your Facebook Likes

In economics and psychology, delay discounting is often used to characterize how individuals choose between a smaller immediate reward and a larger delayed reward. People with higher delay discounting rate (DDR) often choose smaller but more immediate rewards (a "today person"). In contrast, people with a lower discounting rate often choose a larger future rewards (a "tomorrow person"). Since the ability to modulate the desire of immediate gratification for long term rewards plays an important role in our decision-making, the lower discounting rate often predicts better social, academic and health outcomes. In contrast, the higher discounting rate is often associated with problematic behaviors such as alcohol/drug abuse, pathological gambling and credit card default. Thus, research on understanding and moderating delay discounting has the potential to produce substantial societal benefits.

cs.AI↗

arXiv2016

Personalized Emphasis Framing for Persuasive Message Generation

In this paper, we present a study on personalized emphasis framing which can be used to tailor the content of a message to enhance its appeal to different individuals. With this framework, we directly model content selection decisions based on a set of psychologically-motivated domain-independent personal traits including personality (e.g., extraversion and conscientiousness) and basic human values (e.g., self-transcendence and hedonism). We also demonstrate how the analysis results can be used in automated personalized content selection for persuasive message generation.

cs.AI↗

arXiv2015

An Uncertainty-Aware Approach for Exploratory Microblog Retrieval

Although there has been a great deal of interest in analyzing customer opinions and breaking news in microblogs, progress has been hampered by the lack of an effective mechanism to discover and retrieve data of interest from microblogs. To address this problem, we have developed an uncertainty-aware visual analytics approach to retrieve salient posts, users, and hashtags. We extend an existing ranking technique to compute a multifaceted retrieval result: the mutual reinforcement rank of a graph node, the uncertainty of each rank, and the propagation of uncertainty among different graph nodes. To illustrate the three facets, we have also designed a composite visualization with three visual components: a graph visualization, an uncertainty glyph, and a flow map. The graph visualization with glyphs, the flow map, and the uncertainty analysis together enable analysts to effectively find the most uncertain results and interactively refine them. We have applied our approach to several Twitter datasets. Qualitative evaluation and two real-world case studies demonstrate the promise of our approach for retrieving high-quality microblog data.

cs.IR↗

arXiv2015

LDAExplore: Visualizing Topic Models Generated Using Latent Dirichlet Allocation

We present LDAExplore, a tool to visualize topic distributions in a given document corpus that are generated using Topic Modeling methods. Latent Dirichlet Allocation (LDA) is one of the basic methods that is predominantly used to generate topics. One of the problems with methods like LDA is that users who apply them may not understand the topics that are generated. Also, users may find it difficult to search correlated topics and correlated documents. LDAExplore, tries to alleviate these problems by visualizing topic and word distributions generated from the document corpus and allowing the user to interact with them. The system is designed for users, who have minimal knowledge of LDA or Topic Modelling methods. To evaluate our design, we run a pilot study which uses the abstracts of 322 Information Visualization papers, where every abstract is considered a document. The topics generated are then explored by users. The results show that users are able to find correlated documents and group them based on topics that are similar.

cs.IR↗

arXiv1999

Empirically Evaluating an Adaptable Spoken Dialogue System

Recent technological advances have made it possible to build real-time, interactive spoken dialogue systems for a wide variety of applications. However, when users do not respect the limitations of such systems, performance typically degrades. Although users differ with respect to their knowledge of system limitations, and although different dialogue strategies make system limitations more apparent to users, most current systems do not try to improve performance by adapting dialogue behavior to individual users. This paper presents an empirical evaluation of TOOT, an adaptable spoken dialogue system for retrieving train schedules on the web. We conduct an experiment in which 20 users carry out 4 tasks with both adaptable and non-adaptable versions of TOOT, resulting in a corpus of 80 dialogues. The values for a wide range of evaluation measures are then extracted from this corpus. Our results show that adaptable TOOT generally outperforms non-adaptable TOOT, and that the utility of adaptation depends on TOOT's initial dialogue strategies.

cs.CL↗