Toward a More Reliable VGI Quality Assessment Method based on Trustworthiness |
Paper ID : 1071-SMPR-FULL |
Authors: |
امیرمسعود فراتی *1, فرید کریمی پور2 1پونک - انتهای بلوار سردار جنگل - خیابان لادن - خیابان ارکیده - کوچه آیت - پلاک 32 - ط اول 2استادیار دانشکده نقشه برداری و اطلاعات مکانی دانشگاه تهران |
Abstract: |
Abstract: Recent advances in technology makes a great transformation on how geographic information are produced and the phenomenon of VGI appeared, VGI allows people with little knowledge in geographic information to contribute to the creation of maps and other kinds of geographic information so because VGI gathered by individuals with no formal training, the credibility and reliability of VGI is challenging. In this paper we study what kinds of things might contribute to assess trustworthiness of data, and reputation of contributors for VGI. We present a model for analyzing the different factors, and a method for automatically creating the trust and reputation scores. Introduction: Volunteered Geographic Information (VGI) is an approach to crowdsource information about geospatial features around us. Recently, VGI is gaining increasing attention and (web) services relying on it are becoming ubiquitous [1]. because of collaborative environment of VGI that content is created in an arbitrary process and there is no centralized control of the virtual communities involved also the specific nature of VGI that users are non-expert and specifically not geospatial information expert, the ambiguous nature of spatial entities and the large number of users with diverse motivation and backgrounds that any of them have their own perception and conceptualization and cultural background with multifarious spatial enablement so we are faced with a heterogeneous credibility. In particular there are no means of determining the standardized aspect of geospatial information credibility and we need to use VGI specific methods to assess the credibility of this kind of information. A basic method to assess the credibility of volunteer geographic information datasets is comparing it against an authoritative data set that produced professionally, but this method have two major drawbacks, first the authoritative datasets in some places is not available or is very expensive, second this assessment method is not universally valid . So we need to use some alternative methods to assess the credibility of VGI to overcome this drawbacks, we use trust in social networks as a foundation of our alternative credibility assessment approach in fact trust in social networks is a measure of how information produced by some users is relatively valuable to others. Indeed, the more useful information produced by more trusted users. We draw inspiration from the Kessler and extend it by relating data trustworthiness, user reputation and accounting social group reputation to overcome some of limitations of traditional method. Model Overview: The information that users enter to the VGI System reflects their personality and the collection of that information make the users reputation in the VGI system in fact we view reputation as the perception trustworthiness of a person by a community [6] reputation of a person belongs to the community and it depends on many social and psychological factors. Such factor as previous behavior, community perception of person, the capacity of community to sanction bad behavior; and some psychological factors including gender, age, culture, and occupation et cetera. We proposed a model which assert the degree of trust by reputation of users and their social group reputation; in fact reputation helps a Confiding assess how likely is the trustee reliable [4]. Our model assessing the quality of VGI data through a proxy measure: trustworthiness. Trustworthiness is defined [7] as a “bet about the future contingent action of others”, in this sense, trustworthiness is strictly related to the concept of (others’) reputation and social group reputation that we can estimate future contingent action of others by this two parameter, then trustworthiness is associated to each feature and represents the proxy value of data quality Based on five parameters that introduced by Kessler in [5], we can assign a degree of trust to each feature version then all of our feature has degree of trust, this parameter is basis of reputation calculation in fact user reputation depends on the trustworthiness of all the feature version edited by user and is defined as the average of such values: R(u,t)=(∑_(f_i∈F(u,t))▒T(f_i ) )/(|F(u,t)|) Where F(u, t) is the set of all the feature versions edited by user u until time t.[1] Accreditation of the users who we don’t have any history of performance about them and rectification of current user’s credits is done by considering social group reputation because most of human’s action and reactions are influenced by their social elements like their gender their occupation, the environment in which they have grown up and their culture so if we groups the users in such social group by the elements like gender, occupation and age we can rectify uses reputation in fact when a feature created or edited by a user, our anticipation of user reputation is based on his or her social groups reputation. This model is a self-improvement model means the more user statistics more reliable quality assessment so the model improves with the passage of time and increment of users. We can conclude spatial cognition of different social groups of society from this model as the model’s by-product. Conclusion: This paper proposes an approach to include the users’ reputation and their personality characteristics in evaluating the credibility of volunteered geographic information. The initial outcomes of implementing the proposed approach show an improvement in the final results and improved by itself over time. However, grouping and assigning proper initial credibility values are the main concerns to be considered in future. We are also working on new elements that impacts on the spatial ability of users to consider in quality assessment, and finally we are implementing this type of quality assessment on landscape description system. |
Keywords: |
GIS, Volunteer Geographic Information, Data Quality, Trust, Reputation |
Status : Paper Accepted (Poster Presentation) |