Product Recommendations System Survey
Sahil Pathan, Karan Panjwani, Nitin Yadav, Shreyas Lokhande, Bhushan S. Thakare · International Journal of Computer Applications · 2015
Recommendation Systems are used to increase the growth of various online businesses.E-commerce players are utilizing such systems to get high sales.Such systems make use of statistics and data from user behaviour e.g.Purchase history, product ratings.So, decision to display a specific product from a specific category is taken after considering such parameters.In Hyper-Local based services (Locality Based) recommendation systems operate in a challenging environment.Such as, new customers have too much limited information associated, less purchase history, no product ratings etc. Secondly a large retailer have too much categories to choose from.Last, users tends have scattered data-less patterns.In order to handle such information mainly three methods are available: search-based methods, collaborative filtering and cluster models.These methods are more suitable in a vast user base environment.Whereas, in small scale environments a set of customers whose purchased and rated products overlaps with a current user's purchased and rated products are subject to a simple measurements.