Improving searching usability in direct shopping portals

Abstract

E-commerce became very popular in recent years because of the great success achieved by some applications such as online auctions. However, currently the trend is that users of these portals execute more transactions in direct shopping than in online auctions and, probably because of the beginning of many of such portals as auctioning sites, their effectiveness decreases when searching for products in the direct shopping context. In this paper, we introduce a usability-improved fuzzy searching mechanism that allows the specification of uncertain and vague searching preferences to return more relevant results in direct shopping.

A Recommender System Based on a Machine Learning Algorithm for B2C Portals

“…In this experiment, we focus on how the customer is provided with knowledge about the market which is useful to define more realistic preferences. For a more detailed explanation on how the search process is carried out, the reader is referred to [15].…”

“…These issues are consequence of the great amount of data that customers are required to input to the system when specifying a new search. To overcome these problems, in [15] a search and recommender system is proposed that allows the customer to provide the system with a features vector composed of a set of pairs attribute-value which defines the customer's preferences. In this work, we propose a recommender system based on fuzzy logic which is able to work with features vectors like that proposed in [15].…”

“…We have developed a prototype of B2C portal called eZoco [15] aimed at guiding the users in defining their desired product and in helping them to find that product which best matches their expectancies. In this section, we focus on the cataloguing and the search services, as they are the background infrastructure of the proposal devised in this paper.…”

“…to obtain parameters a and d of the trapezoid. For a more detailed explanation on how each data type is valued, the reader is referred to [15]. A variable valuation is then denoted as a 3-tuple…”

ACT 2.0: the next generation of assistive consumer technology research

“…In previous works (Lopez-Lopez et al, 2009b), a centralized architecture called e-ZOCO was developed to give support to an e-commerce application (Miguel and Castro-Schez, 2009). This architecture presents performance problems when the data layer contains lots of data and lots of users are concurrently connected.…”