Search

Given the rapid recent trend of urbanization, a better understanding of how urban infrastructure mediates socioeconomic interactions and economic systems is of vital importance. While the accessibility of location-enabled devices as well as large-scale datasets of human activities, has fueled significant advances in our understanding, there is little agreement on the linkage between socioeconomic status and its influence on movement patterns, in particular, the role of inequality. Here, we analyze a heavily aggregated and anonymized summary of global mobility and investigate the relationships between socioeconomic status and mobility across a hundred cities in the US and Brazil. We uncover two types of relationships, finding either a clear connection or little-to-no interdependencies. The former tend to be characterized by low levels of public transportation usage, inequitable access to basic amenities and services, and segregated clusters of communities in terms of income, with the latter class showing the opposite trends. Our findings provide useful lessons in designing urban habitats that serve the larger interests of all inhabitants irrespective of their economic status., Comment: main manuscript 25 page, 7 Figures. Supplementary material 11 pages, 8 figures

Depressive disorder is one of the most prevalent mental illnesses among the global population. However, traditional screening methods require exacting in-person interviews and may fail to provide immediate interventions. In this work, we leverage ubiquitous personal longitudinal Google Search and YouTube engagement logs to detect individuals with depressive disorder. We collected Google Search and YouTube history data and clinical depression evaluation results from $212$ participants ($99$ of them suffered from moderate to severe depressions). We then propose a personalized framework for classifying individuals with and without depression symptoms based on mutual-exciting point process that captures both the temporal and semantic aspects of online activities. Our best model achieved an average F1 score of $0.77 \pm 0.04$ and an AUC ROC of $0.81 \pm 0.02$., Comment: Machine Learning in Public Health (MLPH) at NeurIPS 2020

Mental health problems among the global population are worsened during the coronavirus disease (COVID-19). How individuals engage with online platforms such as Google Search and YouTube undergoes drastic shifts due to pandemic and subsequent lockdowns. Such ubiquitous daily behaviors on online platforms have the potential to capture and correlate with clinically alarming deteriorations in mental health profiles in a non-invasive manner. The goal of this study is to examine, among college students, the relationship between deteriorating mental health conditions and changes in user behaviors when engaging with Google Search and YouTube during COVID-19. This study recruited a cohort of 49 students from a U.S. college campus during January 2020 (prior to the pandemic) and measured the anxiety and depression levels of each participant. This study followed up with the same cohort during May 2020 (during the pandemic), and the anxiety and depression levels were assessed again. The longitudinal Google Search and YouTube history data were anonymized and collected. From individual-level Google Search and YouTube histories, we developed 5 signals that can quantify shifts in online behaviors during the pandemic. We then assessed the differences between groups with and without deteriorating mental health profiles in terms of these features. Significant features included late-night online activities, continuous usages, and time away from the internet, porn consumptions, and keywords associated with negative emotions, social activities, and personal affairs. Though further studies are required, our results demonstrated the feasibility of utilizing pervasive online data to establish non-invasive surveillance systems for mental health conditions that bypasses many disadvantages of existing screening methods.

This paper describes the SemEval-2020 shared task "Assessing Humor in Edited News Headlines." The task's dataset contains news headlines in which short edits were applied to make them funny, and the funniness of these edited headlines was rated using crowdsourcing. This task includes two subtasks, the first of which is to estimate the funniness of headlines on a humor scale in the interval 0-3. The second subtask is to predict, for a pair of edited versions of the same original headline, which is the funnier version. To date, this task is the most popular shared computational humor task, attracting 48 teams for the first subtask and 31 teams for the second.

Anxiety disorder is one of the world's most prevalent mental health conditions, arising from complex interactions of biological and environmental factors and severely interfering one's ability to lead normal life activities. Current methods for detecting anxiety heavily rely on in-person interviews, which can be expensive, time-consuming, and blocked by social stigmas. In this work, we propose an alternative method to identify individuals with anxiety and further estimate their levels of anxiety using personal online activity histories from YouTube and the Google Search engine, platforms that are used by millions of people daily. We ran a longitudinal study and collected multiple rounds of anonymized YouTube and Google Search logs from volunteering participants, along with their clinically validated ground-truth anxiety assessment scores. We then developed explainable features that capture both the temporal and contextual aspects of online behaviors. Using those, we were able to train models that (i) identify individuals having anxiety disorder with an average F1 score of 0.83 and (ii) assess the level of anxiety by predicting the gold standard Generalized Anxiety Disorder 7-item scores (ranges from 0 to 21) with a mean square error of 1.87 based on the ubiquitous individual-level online engagement data. Our proposed anxiety assessment framework is cost-effective, time-saving, scalable, and opens the door for it to be deployed in real-world clinical settings, empowering care providers and therapists to learn about anxiety disorders of patients non-invasively at any moment in time.

We present a simple yet highly generalizable method for explaining interacting parts within a neural network's reasoning process. First, we design an algorithm based on cross derivatives for computing statistical interaction effects between individual features, which is generalized to both 2-way and higher-order (3-way or more) interactions. We present results side by side with a weight-based attribution technique, corroborating that cross derivatives are a superior metric for both 2-way and higher-order interaction detection. Moreover, we extend the use of cross derivatives as an explanatory device in neural networks to the computer vision setting by expanding Grad-CAM, a popular gradient-based explanatory tool for CNNs, to the higher order. While Grad-CAM can only explain the importance of individual objects in images, our method, which we call Taylor-CAM, can explain a neural network's relational reasoning across multiple objects. We show the success of our explanations both qualitatively and quantitatively, including with a user study. We will release all code as a tool package to facilitate explainable deep learning., Comment: Presented at ICCV, 2021

Nighttime lights satellite imagery has been used for decades as a uniform, global source of data for studying a wide range of socioeconomic factors. Recently, another more terrestrial source is producing data with similarly uniform global coverage: anonymous and aggregated smart phone location. This data, which measures the movement patterns of people and populations rather than the light they produce, could prove just as valuable in decades to come. In fact, since human mobility is far more directly related to the socioeconomic variables being predicted, it has an even greater potential. Additionally, since cell phone locations can be aggregated in real time while preserving individual user privacy, it will be possible to conduct studies that would previously have been impossible because they require data from the present. Of course, it will take quite some time to establish the new techniques necessary to apply human mobility data to problems traditionally studied with satellite imagery and to conceptualize and develop new real time applications. In this study we demonstrate that it is possible to accelerate this process by inferring artificial nighttime satellite imagery from human mobility data, while maintaining a strong differential privacy guarantee. We also show that these artificial maps can be used to infer socioeconomic variables, often with greater accuracy than using actual satellite imagery. Along the way, we find that the relationship between mobility and light emissions is both nonlinear and varies considerably around the globe. Finally, we show that models based on human mobility can significantly improve our understanding of society at a global scale., Comment: 9 pages, 3 figures, presented at the 34th AAAI conference on Artificial Intelligence

Building datasets of creative text, such as humor, is quite challenging. We introduce FunLines, a competitive game where players edit news headlines to make them funny, and where they rate the funniness of headlines edited by others. FunLines makes the humor generation process fun, interactive, collaborative, rewarding and educational, keeping players engaged and providing humor data at a very low cost compared to traditional crowdsourcing approaches. FunLines offers useful performance feedback, assisting players in getting better over time at generating and assessing humor, as our analysis shows. This helps to further increase the quality of the generated dataset. We show the effectiveness of this data by training humor classification models that outperform a previous benchmark, and we release this dataset to the public.

A new clinical literature search engine, called CupQ, is presented. It aims to help clinicians stay updated with medical knowledge. Although PubMed is currently one of the most widely used digital libraries for biomedical information, it frequently does not return clinically relevant results. CupQ utilizes a ranking algorithm that filters non-medical journals, compares semantic similarity between queries, and incorporates journal impact factor and publication date. It organizes search results into useful categories for medical practitioners: reviews, guidelines, and studies. Qualitative comparisons suggest that CupQ may return more clinically relevant information than PubMed. CupQ is available at https://cupq.io/.

Online news has made dissemination of information a faster and more efficient process. Additionally, the shift from a print medium to an online interface has enabled user interactions, creating a space to mutually understand the reader responses generated by the consumption of news articles. Intermittently, the positive environment is transformed into a hate-spewing contest, with the amount and target of incivility varying depending on the specific news website in question. In this paper, we develop methods to study the emergence of incivility within the reader communities in news sites. First, we create a dataset of political news articles and their reader comments from partisan news sites. Then, we train classifiers to predict different aspects of uncivil speech in comments. We apply these classifiers to predict whether a news article is likely to provoke a substantial portion of reader comments containing uncivil language by analyzing only the article's content. Finally, we devise a technique to "read between the lines" --- finding the topics of discussions that an article triggers among its readers without frequent, explicit mentions of these topics in its content., Comment: Peer reviewed at NECO-18 (hosted at International AAAI Conference on Web and Social Media 2018)

Acute Kidney Injury (AKI), a sudden decline in kidney function, is associated with increased mortality, morbidity, length of stay, and hospital cost. Since AKI is sometimes preventable, there is great interest in prediction. Most existing studies consider all patients and therefore restrict to features available in the first hours of hospitalization. Here, the focus is instead on rehospitalized patients, a cohort in which rich longitudinal features from prior hospitalizations can be analyzed. Our objective is to provide a risk score directly at hospital re-entry. Gradient boosting, penalized logistic regression (with and without stability selection), and a recurrent neural network are trained on two years of adult inpatient EHR data (3,387 attributes for 34,505 patients who generated 90,013 training samples with 5,618 cases and 84,395 controls). Predictions are internally evaluated with 50 iterations of 5-fold grouped cross-validation with special emphasis on calibration, an analysis of which is performed at the patient as well as hospitalization level. Error is assessed with respect to diagnosis, race, age, gender, AKI identification method, and hospital utilization. In an additional experiment, the regularization penalty is severely increased to induce parsimony and interpretability. Predictors identified for rehospitalized patients are also reported with a special analysis of medications that might be modifiable risk factors. Insights from this study might be used to construct a predictive tool for AKI in rehospitalized patients. An accurate estimate of AKI risk at hospital entry might serve as a prior for an admitting provider or another predictive algorithm., Comment: In revision

Lifestyles are a valuable model for understanding individuals' physical and mental lives, comparing social groups, and making recommendations for improving people's lives. In this paper, we examine and compare lifestyle behaviors of people living in cities of different sizes, utilizing freely available social media data as a large-scale, low-cost alternative to traditional survey methods. We use the Greater New York City area as a representative for large cities, and the Greater Rochester area as a representative for smaller cities in the United States. We employed matrix factor analysis as an unsupervised method to extract salient mobility and work-rest patterns for a large population of users within each metropolitan area. We discovered interesting human behavior patterns at both a larger scale and a finer granularity than is present in previous literature, some of which allow us to quantitatively compare the behaviors of individuals of living in big cities to those living in small cities. We believe that our social media-based approach to lifestyle analysis represents a powerful tool for social computing in the big data age., Comment: Published at IEEE transactions on Big Data

Nearly all previous work on geo-locating latent states and activities from social media confounds general discussions about activities, self-reports of users participating in those activities at times in the past or future, and self-reports made at the immediate time and place the activity occurs. Activities, such as alcohol consumption, may occur at different places and types of places, and it is important not only to detect the local regions where these activities occur, but also to analyze the degree of participation in them by local residents. In this paper, we develop new machine learning based methods for fine-grained localization of activities and home locations from Twitter data. We apply these methods to discover and compare alcohol consumption patterns in a large urban area, New York City, and a more suburban and rural area, Monroe County. We find positive correlations between the rate of alcohol consumption reported among a community's Twitter users and the density of alcohol outlets, demonstrating that the degree of correlation varies significantly between urban and suburban areas. While our experiments are focused on alcohol use, our methods for locating homes and distinguishing temporally-specific self-reports are applicable to a broad range of behaviors and latent states., Comment: 12 pages, 7 figures, 4-page poster version accepted at ICWSM 2016, alcohol dataset and keywords available in: cs.rochester.edu/u/nhossain/icwsm-16-data.zip

Working adults spend nearly one third of their daily time at their jobs. In this paper, we study job-related social media discourse from a community of users. We use both crowdsourcing and local expertise to train a classifier to detect job-related messages on Twitter. Additionally, we analyze the linguistic differences in a job-related corpus of tweets between individual users vs. commercial accounts. The volumes of job-related tweets from individual users indicate that people use Twitter with distinct monthly, daily, and hourly patterns. We further show that the moods associated with jobs, positive and negative, have unique diurnal rhythms., Comment: 9 pages, 7 figures, 7 tables

Recent research has shown that surprisingly rich models of human activity can be learned from GPS (positional) data. However, most effort to date has concentrated on modeling single individuals or statistical properties of groups of people. Moreover, prior work focused solely on modeling actual successful executions (and not failed or attempted executions) of the activities of interest. We, in contrast, take on the task of understanding human interactions, attempted interactions, and intentions from noisy sensor data in a fully relational multi-agent setting. We use a real-world game of capture the flag to illustrate our approach in a well-defined domain that involves many distinct cooperative and competitive joint activities. We model the domain using Markov logic, a statistical-relational language, and learn a theory that jointly denoises the data and infers occurrences of high-level activities, such as a player capturing an enemy. Our unified model combines constraints imposed by the geometry of the game area, the motion model of the players, and by the rules and dynamics of the game in a probabilistically and logically sound fashion. We show that while it may be impossible to directly detect a multi-agent activity due to sensor noise or malfunction, the occurrence of the activity can still be inferred by considering both its impact on the future behaviors of the people involved as well as the events that could have preceded it. Further, we show that given a model of successfully performed multi-agent activities, along with a set of examples of failed attempts at the same activities, our system automatically learns an augmented model that is capable of recognizing success and failure, as well as goals of peoples actions with high accuracy. We compare our approach with other alternatives and show that our unified model, which takes into account not only relationships among individual players, but also relationships among activities over the entire length of a game, although more computationally costly, is significantly more accurate. Finally, we demonstrate that explicitly modeling unsuccessful attempts boosts performance on other important recognition tasks.

We propose modal Markov logic as an extension of propositional Markov logic to reason under the principle of maximum entropy for modal logics K45, KD45, and S5. Analogous to propositional Markov logic, the knowledge base consists of weighted formulas, whose weights are learned from data. However, in contrast to Markov logic, in our framework we use the knowledge base to define a probability distribution over non-equivalent epistemic situations (pointed Kripke structures) rather than over atoms, and use this distribution to assign probabilities to modal formulas. As in all probabilistic representations, the central task in our framework is inference. Although the size of the state space grows doubly exponentially in the number of propositions in the domain, we provide an algorithm that scales only exponentially in the size of the knowledge base. Finally, we briefly discuss the case of languages with an infinite number of propositions., Comment: 7 pages, Poster presentation at TARK 2013 (arXiv:1310.6382) http://www.tark.org

We are developing a general framework for using learned Bayesian models for decision-theoretic control of search and reasoningalgorithms. We illustrate the approach on the specific task of controlling both general and domain-specific solvers on a hard class of structured constraint satisfaction problems. A successful strategyfor reducing the high (and even infinite) variance in running time typically exhibited by backtracking search algorithms is to cut off and restart the search if a solution is not found within a certainamount of time. Previous work on restart strategies have employed fixed cut off values. We show how to create a dynamic cut off strategy by learning a Bayesian model that predicts the ultimate length of a trial based on observing the early behavior of the search algorithm. Furthermore, we describe the general conditions under which a dynamic restart strategy can outperform the theoretically optimal fixed strategy., Comment: Appears in Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence (UAI2001)