Finding a Trustworthy Service Provider in Online Service Oriented Trust Network

In a service-oriented online social network consisting of service providers and consumers as participants, a service consumer can search trustworthy service providers via the social network between them. In service-oriented Trust network, there is multiple service provider sand multiple paths. Select the best service provider by using trust rate and recommendation of the participants about providers.

In this paper, we are dealing with an algorithm named Gossiping algorithm for finding the best-trusted service provider. Using a request-reply method. Service consumer requested to the provider via their neighbors for service. The provider replied with their recommendation and trust rate. Based on this Consumer find the best-trusted service provider.

Online social networking sites have become very popular and attracting a large number of participants. OSNare being used as a means for a variety of rich activities. Social network as a dedicated website or application which enables users to communicate with each other by posting comments, messages, images, emotions etc (e.g., ebay.com, facebook.com). According to surveys, the usage of OSN is increasing in day by day. Business professionals to analyze customer’s conversations on social networking sites, and as a consequence, provides real-time status updates about their products and services accordingly. In service-oriented environments, social networks can be used as the means for service consumers to look for trustworthy service providers with the assistance of other consumers.

The consumer and provider are unknown to each and no knowledge about their services. For example, at FilmTrust, which is a social networking site for movie recommendations, a recommendation receiver can evaluate the trustworthiness of a recommender via the social network between them. Consider a social network consists of lots of buyers and sellers, it can be used by a buyer to find the most trustworthy seller who sells the product preferred by the buyer[12].In above situations trust is one of the most factors for decision making and mechanisms for evaluating the trustworthiness between participants. Trust is the belief of one participant to another.

A social network can be modeled as a directed graph, Where nodes represent participants, Whereas edges represent relationships of a certain type between them. Figure 1. Service-oriented online social network Fig.1, SC and SP represent service consumer and service provider respectively. Edges as the relation sand their trust rate between each participant. As each participant usually interacts with many other participants. In fig.1 the service consumer and service provider have may be direct path multiple paths exists between them. The service consumer can evaluate the trustworthiness of the provider one based on the trust information between the intermediate participants along the path. This process is called trust propagation and the path with trust information linking the source participant and the target one is called a social trust path[15].

In large scale complex social network, the service consumer finds a good trustworthy service provider is very difficult and time-consuming. The challenging problem is that there is a large number of the service provider and there exist multiple paths between them. We can search the most trustworthy and best service provider, Some of them neglect the recommendation of other participants and trust information in their path. So we proposing a new model for finding the best-trusted service provider. The proposing model, we using the trust information found in the path between provider and consumer and recommendation of another participant about the provider. Our novel search strategies can help deliver better solutions than Other existing methods.

Social Trust Path Selection Small Blue [16] is an online social network constructed for IBM staff. In this system, up to 16 social paths with no more than 6 hops are selected between a source and a target and the shortest path is taken. In this method, it does not consider the important factors such as trust information and the recommendations of other participants. Hang et al. [7] further take the trust between participants into consideration in path selection. In their model, the path with the maximum of propagated trust values is selected as the optimal one. In these methods, some significant influence factors including recommendation roles and social relationships between participants are not taken into account in path selection.

In addition, Guha et al. [14] propose a trust propagation model, where the number of hops in trust propagation is considered in calculating the propagated trust values between a source participant and the target one. Furthermore, in the model of [12], a buyer performs several random walks with a fixed number of hops along with a path from this buyer in the social network to find the ratings of the ending participant to a seller. The degree of confidence of the seller is calculated based on the number of random walk hops, ratings and the number of random walk paths.

Yan Wang and Mehmet [3] propose a method called Heuristic algorithm which is based on the concept of Quality of Trust which includes trust information, recommendation, and relations. In this model find the shortest path between source and target by a forward search and backward search using Dijkstra’s algorithm, and it checks the path to satisfy the QoT. These algorithms have both the efficiency and the quality of identified social trust paths. But it faces some imbalance problem of QoT attributes, which may cause failed feasibility estimation of the path. So it can overcome by proposing another method MFPB-HOSTP[6] which is highly efficient and which give good trust path. This fab-host method is very time-consuming. B. find trustworthy service provider For find the good trustworthy service provider there exist several methods.

A method called Guven [2] which is an incorporates a probabilistic method for trust based on commitment outcomes (based on the interaction between users ) and estimate trust value. This method has more chance to increase the error so accuracy is less because based on prediction techniques. A factor-enrichment-based hybrid trust framework(HT-TRUST) [11] and collaborative techniques [10] are based on opinion mining of the participants. In this methods, the opinions are converted into corresponding mathematical models. In these above applications is mention that, a service consumer can evaluate the trustworthiness of the service provider and can have different trust evaluation criteria in different applications. We have to propose a new method for finding the good trusted service provider by combining the trust path selection methods and recommendation(opinion) of other participants. Propose a new algorithm which gives the new trusted service provider.

Trust network: Trust of a one person to another person is not independent. Trust is dependent on each other as well as transitive. Trust: Trust is the belief of one participant to another based on the past interaction and recommendation of other participants, the future action to be performed by the latter will lead to an expected outcome[15].In a service-oriented network, the flow of trust between participants can be represented in the form of a trust network. The service-oriented trust network can be represented in the form of a directed graph. The provider sand consumers represent the nodes and the relation between edges represents the edges. It contains the trust value to each other.

Trusted systems represent a significant trend in decision support for social networks service provision. Inservice oriented networks trust is based on the feedback on past interactions between members. One participant can give a Trust value to another based on the direct interactions between them. The participants can rate each other Trust increases between members as far as the experience is positive and decreases otherwise. The trust value between two participants is lies between -1 and +1 and 1 represents completely trust to the person, -1 represents completely distrusts and 0 represents neutral trust. Our proposing system finds the most trusted service provider. B. Our Proposed Method We proposing a new method, In our proposed method trust network is considered as the input. Input contains Service providers, Service consumers and thousands of intermediate participants. When a service consumer wants a service, first it sent a request to its neighbors for service and it waits for a reply. The neighbors are not a service provider they forward the request to their corresponding neighbors. Repeat these process until the request reaches to the Service provider. When a request reaches to the service provider, It stops the request forwarding process and sent a reply back to the consumer via their neighbors. The reply contains the star rating of the provider.

When the reply reaches to the neighboring participants it can add the rating of figure 3. The flow of Service request and reply and trust value to SP from neighboring node and send to their neighbors and continue these process until the reply reaches to SC.

Finally, SC estimates the best-trusted service provider. This diagram represents the flow of request messages and replies messages. We have presented a gossiping algorithm for finding a good trusted service provider. This algorithm can search the most trustworthy and best service provider, Some of them neglect the recommendation of other participants and trust information in their path. It takes trust information and recommendation of participants between the provider and consumer.