Los Angeles is ranked the most congested city in the U.S. with a typical half-hour commute taking 81% longer during evening peak periods and 60% longer during the morning peak. These traffic congestions result in a large social and economic detriment and raise serious concern for drivers and transportation agencies. Therefore, increasing ridership of public transportations and hence reducing traffic congestions has been one of the primary objectives for transportation agencies and policymakers. The main objective of this project is to develop reliability analysis system using Deep Learning (DL) techniques that can process massive amounts of 1) GPS trajectories from public transportation vehicles and 2) real-world traffic sensor datasets archived in our data warehouse to predict traffic flow to then enable the forecasting of a variety of performance metrics of public transportation systems.

As mobile phones advance in functionality and capability, they are being used for more than just communication. Increasingly, these devices are being employed as instruments for introspection into habits and situations of individuals and communities. Many of the applications enabled by this new use of mobile phones rely on contextual information. The focus of this work is on one dimension of context, the transportation mode of an individual when outside. We create a convenient (no specific position and orientation setting) classification system that uses a mobile phone with a built-in GPS receiver and an accelerometer. The transportation modes identified include whether an individual is stationary, walking, running, biking, or in motorized transport. The overall classification system consists of a decision tree followed by a first-order discrete Hidden Markov Model and achieves an accuracy level of 93.6% when tested on a dataset obtained from sixteen individuals.

For decades, the Codes of Fair Information Practice have served as a model for data privacy, protecting personal information collected by governments and corporations. But professional data management standards such as the Codes of Fair Information Practice do not take into account a world of distributed data collection, nor the realities of data mining and easy, almost uncontrolled, dissemination. Emerging models of information gathering create an environment where recording devices, deployed by individuals rather than organizations, disrupt expected flows of information in both public and private spaces. We suggest expanding the Codes of Fair Information Practice to protect privacy in this new data reality. An adapted understanding of the Codes of Fair Information Practice can promote individuals’ engagement with their own data, and apply not only to governments and corporations, but software developers creating the data collection programs of the 21st century. To support user participation in regulating sharing and disclosure, we discuss three foundational design principles: primacy of participants, data legibility, and engagement of participants throughout the data life cycle. We also discuss social changes that will need to accompany these design principles, including engagement of groups and appeal to the public sphere, increasing transparency of services through voluntary or regulated labeling, and securing a legal privilege for raw location data.

SYS 2 Controlled Personal Data Stream in Mobile Personal Sensing Min Mun, Shuai Hao, Nilesh Mishra, Katie Shilton, Jeff Burke, Mark Hansen, Ramesh Govindan, Deborah Estrin Mobile Personal Sensing (MPS) is a new kind of participatory sensing where individuals and communities use mobile phones and web-based services to collect and analyze data for use in discovery. While MPS enables people to participate in sensing and analyzing aspects of their lives that were previously invisible, it introduces some constraints due to the inherently intimate nature of the data captured with MPS, which results in concerns for protecting individual privacy, for personal data stream ownership, and for visibility into the web of processing that is used to contextualize and interpret the data. We propose to develop a Personal Data Vault (PDV), a privacy architecture allowing people to control over their data flows over time, with emphasis on controlling the granularities of location data. In addition, we suggest a new method to allow users to use less-granular data for the applications requiring exact location information by partitioning the applications.

Author(s): Banaei-Kashani, Farnoush; Burke, Jeff; Cenizal, Christian; Chen, Suming; Chu, Wesley; Cinnamon, Ian; Dawson, Betta; Denisov, Gleb; Dhanjal, Chandni; Estrin, Deborah; Falaki, Hossein; Govindan, Ramesh; Guan, Zheng; Hansen, Mark; Jia, Nan; Kim, Donnie; Kim, Younghun; Kim, Isaac; Kulinski, Derek; Kutler, Brenden; Longstaff, Brent; Maldonado, Olmo; Mottaghi, Roozbeh; Mun, Min; Nocera, Luciano; Ong, John; Petersen, Nicolai; Ramanathan, Nithya; Reddy, Sasank; Ryder, Jason; Samanta, Vids; Shahabi, Cyrus; Shia, Victor; Shilton, Katie; Shirani-Mehr, Houtan; Srivastava, Mani; Taing, Senglong; Wagmister, Fabian; Wang, Gene; West, Ruth; Whitesell, Kelsey | Abstract: overview poster so no abstract.

Participatory sensing tasks deployed mobile devices to form interactive, participatory sensor networks that enable public and professional users to gather, analyze and share local knowledge. Mobile Personal Sensing (MPS) is a platform for participatory sensing with which users use mobile phones to record and transmit sound, images, location, motion data, and web services to aggregate and interpret the assembled information. The data gathered through MPS is personal, as well as being potentially valuable in many aspects; it quantifies habits, routines, associations, and is easy to mine. However, for these reasons, protecting individual privacy, documenting ownership, and providing visibility of processing are important. We propose Personal Data Vault (PDV), the architecture to support these new design criteria by “auditing” all activities on the data (TraceAudit) and dynamically “re-sampling” data feeds to service providers (Adaptive Filter). The TraceAudit allows the user to track how the data is processed as well as who is using the data in order to provide transparency of data processing and foster a market of “certified” service providers. The adaptive filters govern how the data is sent from PDV to service providers in order to provide a better quality of services with minimal data using two methods: error-tolerant data sampling and anomaly detection.

PEIR, the Personal Environmental Impact Report, is a participatory sensing application that uses location data sampled from everyday mobile phones to calculate personalized estimates of environmental impact and exposure. It is an example of an important class of emerging mobile systems that combine the distributed processing capacity of the web with the personal reach of mobile technology. This paper documents and evaluates the running PEIR system, which includes mobile handset based GPS location data collection, and server-side processing stages such as HMM-based activity classification (to determine transportation mode); automatic location data segmentation into "trips''; lookup of traffic, weather, and other context data needed by the models; and environmental impact and exposure calculation using efficient implementations of established models. Additionally, we describe the user interface components of PEIR and present usage statistics from a two month snapshot of system use. The paper also outlines new algorithmic components developed based on experience with the system and undergoing testing for integration into PEIR, including: new map-matching and GSM-augmented activity classification techniques, and a selective hiding mechanism that generates believable proxy traces for times a user does not want their real location revealed.