Wei-Chen Chen (nickname, Zarcen). I am currently studying in Computer Sciences Dept. in University of Wisconsin-Madison as a graduate student; meanwhile, working as a research assistant supervised by Dr. Michael Gleicher in Vis Lab.
I enjoy developing systems or software which could highly influence daily lives. Aside from Data Visualization that I'm studying with Dr. Michael Gleicher, my broad interests include web application development, machine learning, and ubiquitous computing.
I received my B.S. in computer science from National Taiwan University in 2011. I worked in Smart Home Group of NTU Intelligent Robot Lab and been supervised by Prof. Li-Chen Fu from 2010 to 2011. And involved in IoT and M2M research project as a research assistant in Intel-NTU Center between 2012 and 2014.
By considering the potential merits of behavioral, technological, and organizational changes, our team proposes an ES system named as M2M-based Context-aware Home Energy-Saving System (M-CHESS). The goal of M-CHESS is to optimize context-aware home energy saving and to minimize user's interventions in the determination of ES policies while maintaining the best user comfort requirements through employment of the M2M infrastructure.
Implemented a new approach for Chinese input method in iOS system. We categorized the phonemes in tradition Zhuyin input method into two categories: initial and vowel. Users can intuitively glide between the phonemes and chose their intentional tone by slightly changing their track pattern.
Implemented a dynamic lighting control system that utilizes daylight-harvesting. Integrated with an robust activity rocognition (AR) engine, the system provides proper lighting services to user. It performs daylight detection, inferring user's activity from AR engine, automaticly turning on/off different light set based on the indoor/outdoor illumination and user's activity.
Implemented a HVAC (Heating, Ventilation, and Air Conditioning) control system. The system uses PMV (Predictd Mean Vote, developed by Fanger) indices to design strategies of indoor thermal adjustment. The system's wireless sensors are telosb-architecture running on TinyOS. It utilizes collected contextual data in environment from wireless sensor to infer user's activity to compute accurate user's thermal comfort. Once user actively tunes HVAC system either hotter or colder, it receives the user's feedback to do personal comfort model adaptation.
This was a project in the course Parallel Programming at National Taiwan University. RecastNavigation is a open source project, which is developed by Crisis. The software's goal is to do path finding in a 3D map. However, it takes quite a long time to build the map's NavMesh when the map's size is huge. Thus, we implemented a MPI version program to speed up this bottleneck part by paralleling its mesh-building process.