Search for: All records

In 2011, we proposed PathSim to systematically define and compute similarity between nodes in a heterogeneous information network (HIN), where nodes and links are from different types. In the PathSim paper, we for the first time introduced HIN with general network schema and proposed the concept of meta-paths to systematically define new relation types between nodes. In this paper, we summarize the impact of PathSim paper in both academia and industry. We start from the algorithms that are based on meta-path-based feature engineering, then move on to the recent development in heterogeneous network representation learning, including both shallow network embedding and heterogeneous graph neural networks. In the end, we make the connection between knowledge graphs and HINs and discuss the implication of meta-paths in the symbolic reasoning scenario. Finally, we point out several future directions. 

COVID-19 pandemic has an unprecedented impact all over the world since early 2020. During this public health crisis, reliable forecasting of the disease becomes critical for resource allocation and administrative planning. The results from compartmental models such as SIR and SEIR are popularly referred by CDC and news media. With more and more COVID-19 data becoming available, we examine the following question: Can a direct data-driven approach without modeling the disease spreading dynamics outperform the well referred compartmental models and their variants? In this paper, we show the possibility. It is observed that as COVID-19 spreads at different speed and scale in different geographic regions, it is highly likely that similar progression patterns are shared among these regions within different time periods. This intuition lead us to develop a new neural forecasting model, called Attention Crossing Time Series (ACTS), that makes forecasts via comparing patterns across time series obtained from multiple regions. The attention mechanism originally developed for natural language processing can be leveraged and generalized to materialize this idea. Among 13 out of 18 testings including forecasting newly conrmed cases, hospitalizations and deaths, ACTS outperforms all the leading COVID-19 forecasters highlighted by CDC. 

Trending Topic Detection has been one of the most popular methods to summarize what happens in the real world through the analysis and summarization of social media content. However, as trending topic extraction algorithms become more sophisticated and report additional information like the characteristics of users that participate in a trend, significant and novel privacy issues arise. We introduce a statistical attack to infer sensitive attributes of Online Social Networks users that utilizes such reported community-aware trending topics. Additionally, we provide an algorithmic methodology that alters an existing community-aware trending topic algorithm so that it can preserve the privacy of the involved users while still reporting topics with a satisfactory level of utility. 

Towards the vision of building artificial intelligence systems that can assist with our everyday life, we introduce a proof of concept for a social media privacy "cyborg" which can locally and privately monitor a person's published content and offer advice or warnings when their privacy is at stake. The idea of a cyborg can be more general, as a separate local entity with its own computational resources, that can automatically perform several online tasks on our behalf. For this demonstration, we assume an attacker that can successfully infer user attributes, solely based on what the user has published (topic-based inference). We focus on Social Media privacy and specifically on the issue of exposing sensitive user-attributes, like location, or race, through published content. We built a privacy cyborg that can monitor a user's posted topics and automatically warn them in real time when a sensitive attribute is at risk of being exposed. 

A thorough understanding of social media discussions and the demographics of the users involved in these discussions has become critical for many applications like business or political analysis. Such an understanding and its ramifications on the real world can be enabled through the automatic summarization of Social Media. Trending topics are offered as a high level content recommendation system where users are suggested to view related content if they deem the displayed topics interesting. However, identifying the characteristics of the users focused on each topic can boost the importance even for topics that might not be popular or bursty. We define a way to characterize groups of users that are focused in such topics and propose an efficient and accurate algorithm to extract such communities. Through qualitative and quantitative experimentation we observe that topics with a strong community focus are interesting and more likely to catch the attention of users. 

Abstract Academic researchers, government agencies, industry groups, and individuals have produced forecasts at an unprecedented scale during the COVID-19 pandemic. To leverage these forecasts, the United States Centers for Disease Control and Prevention (CDC) partnered with an academic research lab at the University of Massachusetts Amherst to create the US COVID-19 Forecast Hub. Launched in April 2020, the Forecast Hub is a dataset with point and probabilistic forecasts of incident cases, incident hospitalizations, incident deaths, and cumulative deaths due to COVID-19 at county, state, and national, levels in the United States. Included forecasts represent a variety of modeling approaches, data sources, and assumptions regarding the spread of COVID-19. The goal of this dataset is to establish a standardized and comparable set of short-term forecasts from modeling teams. These data can be used to develop ensemble models, communicate forecasts to the public, create visualizations, compare models, and inform policies regarding COVID-19 mitigation. These open-source data are available via download from GitHub, through an online API, and through R packages. 

Short-term probabilistic forecasts of the trajectory of the COVID-19 pandemic in the United States have served as a visible and important communication channel between the scientific modeling community and both the general public and decision-makers. Forecasting models provide specific, quantitative, and evaluable predictions that inform short-term decisions such as healthcare staffing needs, school closures, and allocation of medical supplies. Starting in April 2020, the US COVID-19 Forecast Hub ( https://covid19forecasthub.org/ ) collected, disseminated, and synthesized tens of millions of specific predictions from more than 90 different academic, industry, and independent research groups. A multimodel ensemble forecast that combined predictions from dozens of groups every week provided the most consistently accurate probabilistic forecasts of incident deaths due to COVID-19 at the state and national level from April 2020 through October 2021. The performance of 27 individual models that submitted complete forecasts of COVID-19 deaths consistently throughout this year showed high variability in forecast skill across time, geospatial units, and forecast horizons. Two-thirds of the models evaluated showed better accuracy than a naïve baseline model. Forecast accuracy degraded as models made predictions further into the future, with probabilistic error at a 20-wk horizon three to five times larger than when predicting at a 1-wk horizon. This project underscores the role that collaboration and active coordination between governmental public-health agencies, academic modeling teams, and industry partners can play in developing modern modeling capabilities to support local, state, and federal response to outbreaks.