University of Twente Student Theses

Login

Understanding of data quality in human sensor web

Sung, Chou-Yen (2010) Understanding of data quality in human sensor web.

[img] PDF
1MB
Abstract:As Volunteered Geographic Information (VGI) evokes many researches about geoinformation collecting, it also causes some problems about its data quality. Because of those VGI contributors without academic or technique background, sometimes those VGI becomes unreliable. Hence, the factors to impede individuals to contribute correct information are needed to be understood. Human sensor web is one typical example of VGI in Zanzibar. Citizens can send SMS to complain about their water problems to human sensor web system. However, because of their culture or their personal judgement about reporting water problem, there are different complaining types. This research used test-retest method, history of water complaining events, water meter data, and the outcome of human sensor web to explore the reason to hinder citizens from contributing correct messages to human sensor web system. One reason to impede citizens send correct message is the information gap between man and woman in one household. Most of mobile phone users in Zanzibar are men, but the one dealing with water collecting job is woman. Even 40% to 50% of women with mobile phones, women still share the ownership of mobile phones with their husbands or relatives. In the process of informing their husbands to send SMS to human sensor web system, it causes the water complaining messages not be sent immediately, information delayed, or wrong messages because of the wrong memory. Mistrust of water provider also causes citizens not willing to use human sensor web system. Because of the water payment issue, unbalanced development between urban and rural area, and no feedback about their complaints, citizens do not trust ZAWA can improve their lives. In this cases, citizens consider human sensor web represents the complaining system of ZAWA. Hence, citizens will not willing to use this system. Local leader (Sheha) plays a crucial role of changing citizens’ willingness to use human sensor web in one community. If sheha support human sensor web system, most of the citizens from that community will be encouraged to use this system. In other case, if sheha rejects to use human sensor web, there will be no one from that community sending complaining messages. Finally, by evaluating the messages from human sensor web system, there are there types of wrong messages be sent by citizens. Wrong typing style, long sentence of SMS, and the SPAM cause those SMS can not be identified and recognizes automatically by system.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:https://purl.utwente.nl/essays/90755
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page