International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395 -0056
Volume: 04 Issue: 01 | Jan -2017
p-ISSN: 2395-0072
www.irjet.net
iCTRE: The Informal community Transformer into Recommendation Engine Vikram R. Raut, Ashish B. Shelke, Akash J. Tripurare, Animesh Kumar Singh
---------------------------------------------------****-----------------------------------------------------Abstract- Presently human is encompassed by a colossal measure of data on the web. That highlights the continuous need of recommendation or suggestion systems in the different areas. Tragically cold start problem is still a critical issue in these systems on new clients and new items. The problem becomes more critical in systems that contain resources that lives too in a matter of seconds like offers on items which remains just for few days (short life resources - SLiR), or news in a news site. From the opposite side social networks are extremely rich with clients' data, shockingly the majority of the proposed social recommender are connected on domain specific social networks like flickers and epinions which are substantially less utilized as a part of the everyday life, since managing General Purpose Social Network (GPSN) like Facebook and Twitter needs to change these GPSN into a valuable source of recommendation dealing with them as row, implicit or unary data. In this work we highlight how iCTRE (Informal community Transformer into Recommendation Engine) addresses this challenge by changing the GPSN into valuable data for recommendation based on middle layer of domain concepts. iCTRE defeats the cold start problem on new clients and items. It has been assessed over Twitter, on new clients, suggesting offers as a sort of SLiR, results demonstrated that iCTRE succeeded in suggesting great offers with 14% of click on suggested offers, which is high contrasted with general open rate in online networking, particularly when we don't have anything about clients and we are suggesting SLiR resources. Introduction Now-a-days recommendation is everywhere; all around us, from choosing a movie, a restaurant as well as
a hotel. Recommendation has become an very essential part of humans daily life. So
collaborative filtering techniques are mostly used to achieve recommendation. In collaborative filtering technics the user is recommended items based on his top N similar users, according to that information the recommendation is provided to that person. If the user consumes the recommended items, this will activate his similarity with others to ultimately follow the mass behavior. Collaborative filtering suffers usually from cold start problem at the level of users and items. Cold start problem happens when new item, for example, has not any actions yet meaning it will not be suggested and not Š 2017, IRJET
|
Impact Factor value: 5.181
|
ISO 9001:2008 Certified Journal
|
Page 915