Intelligent instance selection of data streams for smart sensor applications Magdiel Galan a , Huan Liu a , Kari Torkkola b a Computer Science and Engineering Department, Arizona State University, Tempe, AZ 85287 b Motorola Labs, 2900 South Diablo Way, Tempe, AZ 85282 ABSTRACT The purpose of our work is to mine streaming data from a variety of hundreds of automotive sensors in order to develop methods to minimize driver distraction from in-vehicle communications and entertainment systems such as audio/video devices, cellphones, PDAs, Fax, eMail, and other messaging devices. Our endeavor is to create a safer driving environment, by providing assistance in the form of warning, delaying, or re-routing, incoming signals if the assistance system detects that the driver is performing, or is about to perform, a critical maneuver, such as passing, changing lanes, making a turn, or during a sudden evasive maneuver. To accomplish this, our assistance system relies on maneuver detection by continuously evaluating various embedded vehicle sensors, such as speed, steering, acceleration, lane distance, and many others, combined into representing an instance of the “state” of the vehicle. One key issue is how to effectively and efficiently monitor many sensors with constant data streams. Data streams have their unique characteristics and may produce data that is not relevant or pertinent to a maneuver. We propose an adaptive sampling method that takes advantage of these unique characteristics and develop algorithms that attempt to select relevant and important instances to determine which sensors to monitor and how to provide quick and effective responses to this type of mission critical situations. This work can be extended to many similar sensor applications with data streams. Keywords: Data Mining, Data Streams, Instance Selection, Adaptive Sampling 1. INTRODUCTION The purpose of our work is to develop techniques that will allow us to mine streaming data from hundreds of automotive sensors of various types that continuously monitor a vehicle and its driving environment. Our goal is to capitalize on this information to help us identify when a driver is executing a critical maneuver, in order to develop methods that will aid in minimizing driver distraction from in-vehicle communications systems such as DVD and game players, iPod, cellphones, PDAs, Fax, eMail and other messaging, entertainment, or communication devices. A critical maneuver may include actions such as passing another vehicle, performing a left or right turn, or a sudden evasive maneuver, to name a few. The possibilities of an accident are significantly increased if a driver were to be distracted by any of these devices when performing one of the critical maneuvers. According to statistics by the National Highway Traffic Safety Administration (NHTSA) in 1996, driver distractions contributed to 20-30 percent of all crashes. A more recent study by SAVE-IT (Safety Vehicle Using Adaptive Interface Technology) project funded by the NHTSA showed in their 2003 report that this statistic has increased to 20-50 percent. A contributing factor may be the incorporation of Information and Mobile Technology advances into our vehicles, mainly a result of our demand for constant information. We as individuals maintain busy schedules and often want to maximize the usage of our time, including the time we spend on communication and entertainment in our vehicles. It is not unusual to see drivers having calls or even participating in meetings while driving on the road. Consequently, our vehicles will soon be transformed into a truly mobile information center in order to satisfy our craving for information on anything that might affect our work, business, family, health, or personal finance. Unfortunately, this information- intensive environment significantly increases the cognitive load of a driver, with both the road and communication or entertainment devices competing for the driver’s attention, stretching thin its ability to handle multiple tasks. If not managed properly, a distracted driver could get involved in an accident.