Browse Prior Art Database

Lightweight Characteristic Matching for Ride Sharing Applications

IP.com Disclosure Number: IPCOM000250093D
Publication Date: 2017-May-31
Document File: 4 page(s) / 208K

Publishing Venue

The IP.com Prior Art Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 45% of the total text.

Method and System for Matching Lightweight Characteristics of a Rider with a Driver for Ride Sharing Applications

A method and system is disclosed for matching lightweight characteristics of a rider with a driver for ride sharing applications. The method and system is utilized for matching ride-share drivers with riders by employing objective insights into driver’s behavior and empirical actions along with rider’s character and wishes.

Disclosed is a method and system for matching lightweight characteristics of a rider with a driver for ride sharing applications. Here, the lightweight characteristics can include, but need not be limited to, social factors and preferences of the rider. The method and system is implemented for matching ride-share drivers with riders by employing objective insights into driver’s behavior and empirical actions along with rider’s character and wishes.

The method and system utilizes lightweight social factors and preferences of the rider by automatically collecting data through social media mining and other communication paths to optimize across multiple variables in reaching the rider quickly or selecting a driver with pleasant circumstances.

Firstly, the method and system captures both long-term and real-time information about potential drivers and riders. The information related to drivers and riders can be, but need not be limited to, pick-up point, driving style, talkative/introverted, political learnings to minimize chances of unpleasant interactions, favorite sports and/or teams, celebrities, and/or other points of “small talk commonality”, type of local information provided by the driver, music tastes such as genres, volume levels, current artist playing etc., rider’s destination by a defined time, driver and rider’s psychological profile and other factors which can be categorized to assess commonality or signs of potential personal conflict. This characteristics/information can be gathered explicitly through electronic forms or mined from social media, calendars, conventional communication channels or other known information sources.

Subsequently, the method and system analyzes the information collected from drivers and riders to determine a best fit or combination of multivariant factors between the driver and the rider. Here, the pairing of the driver and the rider is based on weightage applied to the characteristics collected from different sources. For instance, if a user opts for introverted driver, then the importance of sports, celebrities, politics etc., are made zero and an appropriate driver is suggested to the particular rider.

Further, the method and system includes a supervised learning mechanism to refine knowledge based on inputs provided by the rider and the driver. Here the inputs can be measured value, which includes optionally both the electronic forms and sensor reading such as accelerometers etc., and labels which can be the rider’s subjective ratings. Then, the method and system corre...