@conference {351, title = {Estimating Causal Effects of Tone in Online Debates}, booktitle = {International Joint Conference on Artificial Intelligence (IJCAI)}, year = {2019}, abstract = {Statistical methods applied to social media posts shed light on the dynamics of online dialogue. For example, users{\textquoteright} wording choices predict their persuasiveness and users adopt the language patterns of other dialogue participants. In this paper, we estimate the causal effect of reply tones in debates on linguistic and sentiment changes in subsequent responses. The challenge for this estimation is that a reply{\textquoteright}s tone and subsequent responses are confounded by the users{\textquoteright} ideologies on the debate topic and their emotions. To overcome this challenge, we learn representations of ideology using generative models of text.vWe study debates from 4Forums.com and compare annotated tones of replying such as emotional versus factual, or reasonable versus attacking. We show that our latent confounder representation reduces bias in ATE estimation. Our results suggest that factual and asserting tones affect dialogue and provide a methodology for estimating causal effects from text. }, author = {Dhanya Sridhar and Lise Getoor} } @conference {323, title = {A Structured Approach to Understanding Recovery and Relapse in AA}, booktitle = {The Web Conference (WWW)}, year = {2018}, abstract = {

Alcoholism, also known as Alcohol Use Disorder (AUD) is a serious problem affecting millions of people worldwide. Recovery from AUD is known to be challenging and often leads to relapse at various points after enrolling in a rehabilitation program such as Alcoholics Anonymous (AA). In this work, we take a structured approach to understand recovery and relapse from AUD using social media data. To do so, we combine linguistic and psychological attributes of users with relational features that capture useful structure in the user interaction network. We evaluate our models on AA-attending users extracted from the Twitter social network and predict recovery at two different points{\textemdash}90-days and 1 year after the user joins AA, respectively. Our experiments reveal that our structured approach is helpful in predicting recovery in these users. We perform extensive quantitative analysis of different groups of features and dependencies among them. Our analysis sheds light on the role of each feature group and how they combine to predict recovery and relapse. Finally, we present a qualitative analysis of different reasons behind users relapsing to AUD. Our models and analysis are helpful in making meaningful predictions in scenarios where only a subset of features are available and can potentially be helpful in identifying and preventing relapse early.

}, url = {https://github.com/yzhan202/zhang-www18-experiments}, author = {Zhang, Yue and Ramesh, Arti and Golbeck, Jennifer and Dhanya Sridhar and Lise Getoor} } @conference {330, title = {Estimating Causal Effects of Exercise from Mood Logging Data}, booktitle = {ICML Workshop on Causal Machine Learning (CausalML)}, year = {2018}, abstract = {

Mood and activity logging applications empower users to monitor their daily well-being and make informed health choices. To provide users with useful feedback that can improve quality of life, a critical task is understanding the causal effects of daily activities on mood and other wellness markers. In this work, we analyze observational data from EmotiCal, a recently developed mood-logging web application, to explore the effects of exercise on mood.\ We investigate several methodological choices for estimating the conditional average treatment effect, and highlight a novel use of textual data to improve the significance of our results.

}, author = {Dhanya Sridhar and Aaron Springer and Victoria Hollis and Steve Whittaker and Lise Getoor} } @conference {328, title = {Scalable Probabilistic Causal Structure Discovery}, booktitle = {International Joint Conference on Artificial Intelligence (IJCAI)}, year = {2018}, abstract = {

Complex causal networks underlie many real-world problems, from the regulatory interactions between genes to the environmental patterns used to understand climate change. Computational methods seek to infer these casual networks using observational data and domain knowledge. In this paper, we identify three key requirements for inferring the structure of causal networks for scientific discovery: (1) robustness to noise in observed measurements; (2) scalability to handle hundreds of variables; and (3) flexibility to encode domain knowledge and other structural constraints. We first formalize the problem of joint probabilistic causal structure discovery.\ We develop an approach using probabilistic soft logic (PSL) that exploits multiple statistical tests, supports efficient optimization over hundreds of variables, and can easily incorporate structural constraints, including imperfect domain knowledge. We compare our method against multiple well-studied approaches on biological and synthetic datasets, showing improvements of up to 20\% in F1-score over the best performing baseline in realistic settings.

}, url = {https://bitbucket.org/linqs/causpsl/src/master/}, author = {Dhanya Sridhar and Pujara, Jay and Lise Getoor} } @conference {332, title = {Scalable Structure Learning for Probabilistic Soft Logic}, booktitle = {IJCAI Workshop on Statistical Relational AI (StarAI)}, year = {2018}, month = {06/2018}, abstract = {

Statistical relational frameworks such as Markov logic networks and probabilistic soft logic (PSL) encode model structure with weighted first-order logical clauses. Learning these clauses from data is referred to as structure learning. Structure learning alleviates the manual cost of specifying models. However, this benefit comes with high computational costs; structure learning typically requires an expensive search over the space of clauses which involves repeated optimization of clause weights. In this paper, we propose the first two approaches to structure learning for PSL. We introduce a greedy search-based algorithm and a novel optimization method that trade-off scalability and approximations to the structure learning problem in varying ways. The highly scalable optimization method combines data-driven generation of clauses with a piecewise pseudolikelihood (PPLL) objective that learns model structure by optimizing clause weights only once. We compare both methods across five real-world tasks, showing that PPLL achieves an order of magnitude runtime speedup and AUC gains up to 15\% over greedy search.

}, author = {Varun Embar and Dhanya Sridhar and Golnoosh Farnadi and Lise Getoor} } @conference {sridhar:akbc17, title = {Using Noisy Extractions to Discover Causal Knowledge}, booktitle = {NIPS Workshop on Automated Knowledge Base Construction}, year = {2017}, author = {Dhanya Sridhar and Pujara, Jay and Lise Getoor} } @article {sridhar:bioinformatics16, title = {A Probabilistic Approach for Collective Similarity-based Drug-Drug Interaction Prediction}, journal = {Bioinformatics}, volume = {32}, year = {2016}, chapter = {3175--3182}, abstract = {

MOTIVATION: As concurrent use of multiple medications becomes ubiquitous among patients, it is crucial to characterize both adverse and synergistic interactions between drugs. Statistical methods for prediction of putative drug-drug interactions (DDIs) can guide in vitro testing and cut down significant cost and effort. With the abundance of experimental data characterizing drugs and their associated targets, such methods must effectively fuse multiple sources of information and perform inference over the network of drugs.

RESULTS: We propose a probabilistic approach for jointly inferring unknown DDIs from a network of multiple drug-based similarities and known interactions. We use the highly scalable and easily extensible probabilistic programming framework Probabilistic Soft Logic We compare against two methods including a state-of-the-art DDI prediction system across three experiments and show best performing improvements of more than 50\% in AUPR over both baselines. We find five novel interactions validated by external sources among the top-ranked predictions of our model.

AVAILABILITY AND IMPLEMENTATION: Final versions of all datasets and implementations will be made publicly available.

CONTACT: dsridhar@ucsc.edu.

}, author = {Dhanya Sridhar and Shobeir Fakhraei and Lise Getoor} } @conference {sridhar:kddws16, title = {Joint Probabilistic Inference of Causal Structure}, booktitle = {KDD Workshop on CD}, year = {2016}, abstract = {

Causal directed acyclic graphical models (DAGs) are powerful reasoning tools in the study and estimation of cause and effect in scientific and socio-behavioral phenomena. In many domains where the cause and effect structure is unknown, a key challenge in studying causality with DAGs is learning the structure of causal graphs directly from observational data. Traditional approaches to causal structure discovery are categorized as constraint-based or score-based approaches. Score-based methods perform greedy search over the space of models whereas constraint-based methods iteratively prune and orient edges using structural and statistical constraints. However, both types of approaches rely on heuristics that introduce false positives and negatives. In our work, we cast causal structure discovery as an inference problem and propose a joint probabilistic approach for optimizing over model structures. We use a recently introduced and highly efficient probabilistic programming framework known as Probabilistic Soft Logic (PSL) to encode constraint-based structure search. With this novel probabilistic approach to structure discovery, we leverage multiple independence tests and avoid early pruning and variable ordering. We compare our method to the notable PC algorithm on a well-studied synthetic dataset and show improvements in accuracy of predicting causal edges.

}, author = {Dhanya Sridhar and Lise Getoor} } @conference {sridhar:uaiws16, title = {Probabilistic Inference for Causal Structure Discovery}, booktitle = {UAI Workshop on Causation}, year = {2016}, author = {Dhanya Sridhar and Lise Getoor} } @conference {sridhar:acl15, title = {Joint Models of Disagreement and Stance in Online Debate}, booktitle = {Annual Meeting of the Association for Computational Linguistics (ACL)}, year = {2015}, author = {Dhanya Sridhar and Foulds, James and Walker, Marilyn and Huang, Bert and Lise Getoor} } @conference {3, title = {Collective Stance Classification of Posts in Online Debate Forums}, booktitle = {ACL Joint Workshop on Social Dynamics and Personal Attributes in Social Media}, year = {2014}, author = {Dhanya Sridhar and Lise Getoor and Walker, Marilyn} } @conference {sridhar:baylearn14, title = {Collective classification of stance and disagreement in online debate forums}, booktitle = {Bay Area Machine Learning Symposium (BayLearn)}, year = {2014}, author = {Dhanya Sridhar and Foulds, James and Huang, Bert and Walker, Marilyn and Lise Getoor} }