1 Introduction

In this work, a variant to self-organizing maps based on Kohonen’s algorithm is proposed in order to analyze a real B2C websites. The Kohonen algorithm is a neural network proposed by Kohonen, which maps a distribution of vectors of any dimension into a lower-dimensional space, generally one or two, while maintaining a high degree of topological ordering, or neighborhood preservation. It is widely adopted to visualize high-dimensional data sets and to mine online data [1–4]. Social networks such as eBay.com, Amazon.com and Taobao.com (the largest online shopping platform in China) provide forums for consumer ratings, evaluations and advice on users. Consumers using Taobao.com spent USD 180 billion in 2014, and 94 % of their purchases were shared with others [5]. Although social commerce has become an important topic for many researchers, previous studies of social commerce have generally been developed from the literature of e-commerce. Social commerce is online business that combines e-commerce with social media (e.g., twitter) and social networking to accomplish business goals, functions and behaviors. Qu defined social commerce as online business activities initiated via social media which entails business transactions through either social media (e.g., Taobao.com) or other e-commerce sites. Social commerce enables the use of various social technologies to improve the shopping experience for customers. Smart phones are now the most commonly used tools in social commerce transactions. Social commerce adoption is regarded by the computer and information industries as the ‘future generation’ of digital business because of their convenience in social interaction and Internet login [6]. Since smart phones and tablet computers are widely adopted in social interaction and electronic commerce, wearable devices, as the extended or future generation of digital devices, should provide better experiences of social interaction. Therefore, understanding the factors impacting the access of social websites can lead to better academic integrity and market applications. One of the main drawbacks associated with the classical Kohonen’s algorithm results from the prerequisite the adequate size of the output layer in advance [7, 8]. In this work, we extended the existing Kohonen’s model, which allows shortening the time of training with regard to other approaches that also try to solve the problem of the optimal size of SOM, e.g., the growing hierarchical self-organizing maps algorithm [9].



برچسب ها

مطالب پیشنهادی ما

دیدگاه های شما

دیدگاهتان را بنویسید

نشانی ایمیل شما منتشر نخواهد شد. بخش‌های موردنیاز علامت‌گذاری شده‌اند *

آخرین دیدگاه‌ها