Search

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

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., Peer reviewed at NECO-18 (hosted at International AAAI Conference on Web and Social Media 2018)

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., 9 pages, 7 figures, 7 tables

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., 7 pages, Poster presentation at TARK 2013 (arXiv:1310.6382) http://www.tark.org

This Dagstuhl seminar brought together researchers with a wide range of interests and backgrounds related to plan and activity recognition. It featured a substantial set of longer tutorials on aspects of plan and activity recognition, and related topics and useful methods, as a way of establishing a common vocabulary and shared basis of understanding. Building on this shared understanding, individual researchers presented talks about their work in the area. There were also panel discussions which addressed questions about how to best foster progress in the field --- specifically how to improve our ability to compare different plan and activity recognition algorithms --- and address the question of whether to assume rationality in the modeled agents (a question that is of great concern in many fields at this time). This report presents a summary of the talks and discussions at the seminar.

The papers in this proceedings present the latest advances in the field of automated planning and scheduling The accepted papers cover the entire range of planning and scheduling and involve 22 specific research fields, with an emphasis in planning and scheduling under uncertainty, search algorithms, domain independent classical planning, robot planning, and learning for planning and scheduling.

From 12.06.05 to 17.06.2005 the Dagstuhl Seminar 05241 ``Synthesis and Planning'' was held in the International Conference and Research Center (IBFI), Schloss Dagstuhl. During the seminar, several participants presented their current research, and ongoing work and open problems were discussed. Abstracts of the presentations given during the seminar as well as abstracts of seminar results and ideas are put together in this paper. The first section describes the seminar topics and goals in general. Links to extended abstracts or full papers are provided, if available.

In 1998, the international planning community was invited to take part in the first planning competition, hosted by the Artificial Intelligence Planning Systems Conference, to provide a new impetus for empirical evaluation and direct comparison of automatic domain-independent planning systems. This article describes the systems that competed in the event, examines the results, and considers some of the implications for the future of the field.

The difficulty of finding information on the World Wide Web by browsing hypertext documents has led to the development and deployment of various search engines and indexing techniques. However, many information-gathering tasks are better handled by finding a referral to a human expert rather than by simply interacting with online information sources. A personal referral allows a user to judge the quality of the information he or she is receiving as well as to potentially obtain information that is deliberately not made public. The process of finding an expert who is both reliable and likely to respond to the user can be viewed as a search through the net-work of social relationships between individuals as opposed to a search through the network of hypertext documents. The goal of the REFERRAL WEB Project is to create models of social networks by data mining the web and develop tools that use the models to assist in locating experts and related information search and evaluation tasks.

The Fourth International Workshop on Nonmonotonic Reasoning brought together active researchers in nonmonotonic reasoning to discuss current research, results, and problems of both theoretical and practical natures. There was lively discussion on a number of issues, including future research directions for the field.

This paper proposes the utilization of randomized backtracking within complete backtrack search algorithms for propositional satisfiability (SAT). In recent years, randomization has become pervasive in SAT algorithms. Incomplete algorithms for SAT, for example the ones based on local search, often resort to randomization. Complete algorithms also resort to randomization. These include state-of-the-art backtrack search SAT algorithms that often randomize variable selection heuristics. Moreover, it is plain that the introduction of randomization in other components of backtrack search SAT algorithms can potentially yield new competitive search strategies. As a result, we propose a stochastic backtrack search algorithm for SAT, that randomizes both the variable selection and the backtrack steps of the algorithm. In addition, we relate randomized backtracking with a more general form of backtracking, referred to as unrestricted backtracking. Finally, experimental results for different organizations of randomized backtracking are described and compared, providing empirical evidence that the new search algorithm for SAT is a very competitive approach for solving hard real-world instances.

In this position paper we raise the question whether lifted inference can be performed by 'lifting' the probabilistic model rather than the inference algorithm. We show that this is indeed possible using so-called count models, at least for the common examples used within the exact lifting literature. The challenge is to turn this into a general approach. ispartof: International Workshop on Statistical Relational AI location:Catalina Island, USA date:18 Aug - 18 Aug 2012 status: published

Trust plays a key role in social interactions. Explicitly modeling trust is therefore an important aspect of social network analysis in settings such as reputation management systems, recommendation systems, and viral marketing. Within the social sciences, trust is known to depend on network structure, context, individual actors’ attributes, and group memberships and affiliations. Furthermore, trust is often measured quantitatively, according to degrees of trust, rather than as a binary indicator. In this paper, we propose trust modeling as a rich challenge for statistical relational learning (SRL). Additionally, we show that probabilistic soft logic (PSL) is particularly well-suited for this problem. PSL, like many SRL languages, provides an intuitive framework for capturing the relational aspects of trust modeling, while its soft truth values easily accommodate varying strengths of trust. We model various sociological theories of trust in PSL and experimentally compare the resulting PSL programs to existing trust prediction methods, demonstrating the ease of model development and showing that these interpretable first-order logic models produce results of competitive quality. ispartof: pages:1-8 ispartof: International Workshop on Statistical Relational AI pages:1-8 ispartof: International Workshop on Statistical Relational AI location:Catalina Island, California, USA date:18 Aug - 18 Aug 2012 status: published

We propose a partially informed depth-first search algorithm to cope with the Job Shop Scheduling Problem with makespan minimization. The algorithm is built from the well-known P. Brucker's branch and bound algorithm. We improved the heuristic estimation of Brucker's algorithm by means of constraint propagation rules and so devised a more informed heuristic which is proved to be monotonic. We conducted an experimental study across medium and large instances. The results show that the proposed algorithm reaches optimal solutions for medium instances taking less time than branch and bound and that for large instances it reaches much better lower and upper bounds when both algorithms are given the same amount of time, This research has been supported by the Spanish Ministry of Science and Education under research project MEC-FEDER TIN2007-67466-C02-01 and by the Principality of Asturias under grant FICYT-BP09105.

We propose an integration of a fragment or propositional dynamic logic with an epistemic logic so as to represent logically the interactions between action and knowledge that are fundamental to planning order incomplete knowledge and partial observability. This logic, called EDL, enables us representing incomplete knowledge, nondeterministic actions, observations, sensing actions and conditional plans; it also enables a logical expression of several frequently made assumptions about the nature of the problem such as determinism, full observability, unobservability, or pure knowledge gathering. Plan verification corresponds to testing the validity of a given formula of this logic. The allowed plans are conditional, and a key point of our framework is that a plan is meaningful if the branching conditions bear on the knowledge of the agent only, and not on the real world (to which tha agent may not have access); this leads us considering “plans that reason” which may contain branching conditions referring to implicit knowledge to be evaluated at execution time.