User review is a crucial component of open mobile app markets such as the Google Play Store. How do we summarize hundreds or thousands of user reviews and identify the reasons of their likes/dislikes? Unfortunately, beyond simple summaries such as histograms of user ratings, there are few analytic tools that can provide insights into user reviews. In this project, we analyze tens of millions user ratings and comments in mobile app markets at three different levels of detail. Our system is able to (a) discover inconsistencies in reviews; (b) identify reasons why users like or dislike a given app, and provide an interactive, zoomable view of how users' reviews evolve over time; and (c) provide valuable insights into the entire app market, identifying major user concerns and preferences of different types of apps. We evaluated our techniques on a large dataset of 32 GB consisting of over 13 million user reviews of a total 171 thousand Android apps in the Google Play Store. We discuss how the techniques presented herein can be deployed to help a mobile app market operator such as Google Play as well as individual app developers and end-users.
See our KDD 2013 paper on this project.

While prior studies have provided us with an initial understanding of people’s location-sharing privacy preferences, they have been limited to Western countries and have not investigated the impact of the granularity of location disclosures on people’s privacy preferences. We report findings of a three-week comparative study collecting location traces and location-sharing preferences from two comparable groups in the U.S. and China. Results of the study shed further light on the complexity of people’s location-sharing privacy preferences and key attributes influencing willingness to disclose locations to others and to advertisers. While our findings reveal many similarities between U.S. and Chinese participants, they also show interesting differences, such as differences in willingness to share location at ‘home’ and at ‘work’ and differences in the granularity of disclosures people feel comfortable with. We conclude with a discussion of implications for the design of location-sharing applications and location-based advertising.
Paper will be appear in PUC journal soon.

In this project, we explore and compare people’s different behaviors and preferences in two kinds of location sharing scenarios, i.e. social-driven location sharing and purpose-driven location sharing. The popularity of micro-blogging has made general-purpose information sharing a pervasive phenomenon. This trend is now impacting location sharing applications (LSAs) such that users are sharing their location data with a much wider and more diverse audience. We describe this as social-driven sharing, distinguishing it from past examples of what we frame as purpose-driven location sharing. We study the differences between these two types of sharing by conducting a comparative two-week study with nine participants. We found significant differences in terms of users’ decisions about what location information to share, their privacy concerns, and how privacy-preserving their disclosures were.

Karen Tang, Jialiu Lin and Jason Hong, “Rethinking Location Sharing: Exploring the Implications of Social-Driven vs. Purpose-Driven Location Sharing”, Proc. Of the 12th ACM International Conference on Ubiquitous Computing, Copenhagen, Denmark, 2010

Many existing location sharing applications provide coordinate-based location estimates and display them on a map. However, people use a rich variety of terms to convey their location to others, such as “home”, “Starbucks”, or even “the bus stop near my apartment”. Our long-term goal is to create a system that can automatically generate useful place names based on real-time context. Towards this end, we conducted a series of user studies to understand people’s preferences for place naming.

Based on the results, we propose a hierarchical classification of place naming methods. We further conclude that people’s place naming preferences are complex and dynamic, but fairly predictable by using machine learning techniques. Our fndings also suggested that two factors, people’s routine and their willingness to share location information, strongly influence the way people name a place. The new findings provide important implications for location sharing applications and other location based services.

Jialiu Lin, Guang Xiang, Jason Hong and Norman Sadeh, “Modeling People’s Place Naming Preferences in Location Sharing”, Proc. Of the 12th ACM International Conference on Ubiquitous Computing,  Copenhagen, Denmark, 2010

Urban sensing, participatory sensing, and user activity recognition can provide rich contextual information for mobile applications, such as mobile social networking and location-based services. However, continuously capturing this contextual information on mobile devices consumes a huge amount of energy. We proposed a novel framework for an Energy Efficient Mobile Sensing System (EEMSS). EEMSS uses hierarchical sensor management strategies to recognize user states and detect state transitions. By powering only a minimum set of sensors and using appropriate sensor duty cycles EEMSS significantly improves device battery life. We also implement a mobile user state recognition system that automatically identifies a set of users’ daily activities in real time. Our evaluation showed that our approach increases the device battery life by more than 75% while maintaining both high accuracy and low latency in recognizing transitions between end-user activities.

This work was collaborated with Nokia Research Center at Palo Alto, CA and USC.

Yi Wang, Jialiu Lin, Murali Annavaram, Quinn A. Jacobson, Jason Hong, Bhaskar Krishnamachari and Norman Sadeh, “A Framework of Energy Efficient Mobile Sensing for Automatic User State Recognition,” Proc. of the 7th Annual International Conference on Mobile Systems, Applications and Services, Krakow, Poland, June 22-25, 2009

Planning and attending social events are a part of our everyday lives. However, current software does not support all the activities surrounding social events. In particular, while there are many web‐based social event sites that are good for planned events (such as evdb.com, zvents.com, upcoming.org, eventful.com, and SonicLiving), there is little support for ad hoc and impromptu events, for example events that are happening right now (e.g. “free food now”) or later in the day (e.g. “bowling later today?” or “dinner in 15 minutes?”). There is also another implicit issue of this problem, namely that these web sites were designed with the expectation that people would interact with them via desktop computers. Since people are not always at their desktops, this constraint limits spur of the moment decisions to find and attend events. Similarly, constraining ourselves to the desktop computer limits how we share events with friends.

To address this problem, we are developing WhisperMobile, a mobile social application that facilitates creating, finding, and sharing impromptu social events. WhisperMobile lets people create new impromptu events as well as find nearby events that others have posted, and then organize friends to go to that event.

A number of mobile applications have emerged that allow users to locate one another. However, people have expressed concerns about the privacy implications associated with this class of software, which suggests that these kinds of applications can only be broadly adopted if these concerns are adequately addressed.

Locaccino, developed by Mobile Commerce Lab, is an application that enables cell phone and laptop users to selectively share their locations with others (e.g. friends, family, and colleagues). Users interact with their devices to inquire about the locations of other users and to create privacy and obfuscation policies that govern how their own location information may be accessed.

The objective of our work has been to better understand people’s attitudes and behaviors towards privacy as they interact with such an application, and to explore technologies that empower users to more effectively and efficiently specify their privacy preferences (or “policies”). These technologies include user interfaces for specifying rules and auditing disclosures, and also include machine learning techniques to refine user policies based on their feedback.