Anda di halaman 1dari 2

Assignment Achal Kumar Prasad: PhD/MGT/1001 Submitted to: Dr. vidushi Sharma & Dr.

Manisha Sharma Analysis of paper: Using Social Network Analysis as a strategy for E-commerce Recommendation Yonhong et.al. With reference to My Research Project Aiming to magnify the application of WebBased Social Network in Business and Socio-economical Purposes: Background: Being an active member at orkut and face book I was always amazed with the effectiveness of web-based social networks in globalizing (other than our local or easy to reach networks of people) our relationships. Thats again realized as a source of finding a like minded or a person of any context. Now there are hundreds of web-based social networks on internet and networking sites almost cover 80% (observed) of surfing time of people worldwide. So many organizations (of all types) are taking interest in exploiting these networks for their own benefits like advertising, finding customers and employees, market research etc. Using a social network as mentioned above will be more effective if we understand the type, topology and evolution of a network. Earlier these networks were studied as any other networks (computer based or artificial) but soon researchers realized that these are not controlled or well defined still they follow a pattern as per their relationships. So they came up with the concept of scale free networks whose topologies was developed using power law algorithm as they can change with the networks with the relationship type. Thereafter there has been much interest among the researchers of many fields such as economics (countrys trade relations, businesses, internet technologies (website linkages), medical sciences (to trap the network of epidemics) and in some unique analyses like who should be the next President and CEO. The Paper: published as proceedings at PASIC (Pacific Asia Conference on Information Systems) 2009 reports the work done by Yonhong et.al group of professors from different universities of china. It illustrates the concepts of social network analysis and how it can be employed to make better recommendations in e-commerce context with applications and scope of further research. Coverage: Role of Recommenders agent in e-commerce and social networks

As defined by (Xiao et al) recommenders agents elicit the interests of preferences of consumers for products, either explicitly or implicitly, and make recommendations accordingly. So recommender agents enable customized recommendations and creation of a new online store personally designed for each customer. Social network analysis (SNA) has attracted a lot of attention in recent years (Jennifer et al. 2005; Ortega et al. 2008; Otte et al. 2002). Different from variable analysis and topological analysis, which focus on the concepts and attributes of objects, SNA addresses the structure and relationships of objects. SNA sheds light on discovering underlying relationships of people. Take two unknown people and two friends for example. There may be no significant differences in their attributes. However, the relationships between two unknown people and a pair of friends are much different. In recommendation context, the interests of two friends tend to be related to some extent. This kind of information can be used to improve the recommendation results. Measures and applications from network analysis: Development in studies of social networks has given us many measures to define and analys the social networks. Some important concepts and their applications have been discussed in this paper like category of networks, types of centralities and cohesive properties also with the representative examples of use of k-group. These researches are still in primitive stage and for improving the performance of recommendation we have to find out the capabilities of social networks in finding opinion leader nodes and cohesive groups. Directions for further research: Researcher have provided dimensions for future research which are in ways of a. Extracting data from social networks internally and externally b. Integrating with other recommending agents c. Advances in network analysis to provide insights in mining relationships of customers

Anda mungkin juga menyukai