Brent Hecht is an assistant professor of computer science and engineering at the University of Minnesota. With interests that lie at the intersection of human–computer interaction, geography, and big data, his research centers on the relationship between big data and human factors such as culture. A major focus of his work involves volunteered geographic information and its application in location-aware technologies. Dr. Hecht received a PhD in computer science from Northwestern University, a master’s degree in geography from UC Santa Barbara, and dual bachelor’s degrees in computer science and geography from Macalester College. He was a keynote speaker at WikiSym – the premiere conference on wikis and open collaboration – and has received awards for his research at top-tier publication venues in human-computer interaction and geography (e.g., ACM CHI, COSIT). He has collaborated with Google Research, Xerox PARC, and Microsoft Research, and his work been featured in the MIT Technology Review, New Scientist, AllThingsDigital, and various international TV, radio, and Internet outlets.
My current research has two overlapping foci, both of which fall in the domain of social computing and have a heavy emphasis on large-scale analytics. My first focus involves the mining and application of diverse cultural perspectives in user-generated content (UGC). Through the use of machine learning and link analysis, I have found that UGC reflects the cultural diversity of its contributors to a previously unidentified extent, and that this diversity has important implications for Web users and existing UGC-based technologies. For instance, I have shown that according to the supposedly neutral information in the Japanese Wikipedia, Japan is seven times more important to all of encyclopedic world knowledge than any other country in the world. Similar patterns were present in nearly all of the 25 Wikipedia language editions examined. Moreover, I have demonstrated that these cultural biases affect the hundreds of Wikipedia-based technologies that have been developed (e.g., semantic relatedness measures, a family of algorithms widely-used in information retrieval and natural language processing). In effect, these algorithms adopt the points of view of their Wikipedia knowledge base and their output can be highly varied depending on the language edition used.
My second major area of interest looks at the fundamental properties of volunteered geographic information (VGI) such as geotagged photos, Wikipedia articles about places, and geographically referenced tweets. This research has demonstrated, for instance, that users of social media frequently input non-geographic information into the location fields of their user profiles (e.g. fake places like “in Justin Bieber’s heart”) and that this information can easily fool popular systems that convert place names into latitude and longitude coordinates. Implications include “misplaced” users and content in social location-aware technologies and lower-level effects on systems and algorithms that take as input volunteered geographic information.
Another of my projects on geographic UGC involved investigating the extent to which geographic UGC represents local knowledge as opposed to that contributed by tourists or power users in a given social media community. Examining millions of Flickr photos and many language editions of Wikipedia, I showed that Flickr is a far more “local” repository and highlighted a number of means by which developers of social computing systems can engender a higher percentage of local knowledge. Finally, I have also done basic research in this area. For instance, I have shown that the First Law of Geography – that everything is related to everything else, but near things are more related than distant things – has been implicitly encoded in Wikipedia, validating the law in the largest repository of general knowledge in existence.
Hecht, B., Carton, S., Quaderi, M., Schöning, J., Raubal, M., Gergle, D., Downey, D. Explanatory Semantic Relatedness and Explicit Spatialization for Exploratory Search. Proceedings of ACM SIGIR 2012. New York: ACM Press.
Hecht, B., Hong, L., Suh, B. and Chi, E.H. (2011). Tweets from Justin Bieber’s Heart: The Dynamics of the “Location” Field in User Profiles. Proceedings of the ACM Conference on Human Factors in Computing Systems (CHI 2011), pp. 237-246. New York: ACM Press.
Hecht, B. and Gergle, D. (2010). The Tower of Babel Meets Web 2.0: User-Generated Content and Its Applications in a Multilingual Context. Proceedings of the ACM Conference on Human Factors in Computing Systems (CHI 2010), pp. 291–300. New York: ACM Press. (Best Paper Award)
Hecht, B. and Gergle, D. (2010). On The “Localness” of User-Generated Content. Proceedings of the ACM Conference on Computer Supported Cooperative Work (CSCW 2010), pp. 229-232 (short paper). New York: ACM Press.
Bao, P., Hecht, B., Carton, S., Quaderi, M., Horn, M. and Gergle, D. (2012). Omnipedia: Bridging the Wikipedia Language Gap. Proceedings of the ACM Conference on Human Factors in Computing Systems (CHI 2012). New York: ACM Press.