ReputationPro: The Efficient Approaches to Contextual Transaction Trust Computation in E-Commerce Environments

TL;DR

This paper introduces ReputationPro, a new model for computing contextual transaction trust in e-commerce, using a trust vector based on product, price, and time, supported by efficient algorithms and data structures for large-scale data.

Contribution

ReputationPro is the first model to incorporate context dimensions into trust evaluation, with novel algorithms and data structures for efficient computation and storage.

Findings

CMK-tree outperforms existing methods in efficiency

CMK-treeRS reduces storage while maintaining performance

Experimental results validate the effectiveness of the proposed approaches

Abstract

In e-commerce environments, the trustworthiness of a seller is utterly important to potential buyers, especially when the seller is unknown to them. Most existing trust evaluation models compute a single value to reflect the general trust level of a seller without taking any transaction context information into account. In this paper, we first present a trust vector consisting of three values for Contextual Transaction Trust (CTT). In the computation of three CTT values, the identified three important context dimensions, including product category, transaction amount and transaction time, are taken into account. In particular, with different parameters regarding context dimensions that are specified by a buyer, different sets of CTT values can be calculated. As a result, all these values can outline the reputation profile of a seller that indicates the dynamic trust levels of a seller in different product categories, price ranges, time periods, and any necessary combination of them. We term this new model as ReputationPro. However, in ReputationPro, the computation of reputation profile requires novel algorithms for the precomputation of aggregates over large-scale ratings and transaction data of three context dimensions as well as new data structures for appropriately indexing aggregation results to promptly answer buyers' CTT requests. To solve these challenging problems, we then propose a new index scheme CMK-tree. After that, we further extend CMK-tree and propose a CMK-treeRS approach to reducing the storage space allocated to each seller. Finally, the experimental results illustrate that the CMK-tree is superior in efficiency for computing CTT values to all three existing approaches in the literature. In addition, though with reduced storage space, the CMK-treeRS approach can further improve the performance in answering buyers' CTT queries.