The Study On The Brand Structure Assessment Based On BP Neural Network

With the increasingly competitive market, brand has become one of the most strategic assets for corporate sustainable development. Consequently, the fierce competitions ask for the effective development of brand portfolio strategy, as well as s a reasonable brand structure to integrate the brand resources. Since unreasonable brand structure will cause negative impacts on consumer cognition, brand structure has become the key issue of brand integration process, and enterprises need work out a suitable structure to gain core competence. Therefore, the choice of an effective brand structure assessment method is of utmost importance to the brand integration. Under such circumstances, this paper presents a comprehensive assessment method based on BP neural network. The purpose of the study is to discuss the key factors which play the most important role in building brand structure, with both the quantitative and qualitative research. It aims to set up a scientific evaluation system for brand structure assessment, providing scientific and effective basis for corporations. The main research of this article includes two aspects, one of which is to form the index series for evaluation based on the study of factors related to establishment of evaluation system, the other is the application of assessment model with Matlab neural network tools.The article first reviewed the owned literatures and summarized existing research by means of source analysis. Then, the key factors were in-depth analyzed within three perspectives- from consumer, market and enterprise.After the founding of index portfolio, the paper chose the hair care industry as the empirical field, and several representative enterprises were selected as the training samples. Then some secondary data was collected through document retrievals, questionnaire has been designed to survey consumer recognition, and experts were invited to evaluate brand structure of sample firms. With the sample data, the assessment model was built based on the Matlab neural network tools. It has been shown from the simulink practice that this model was able to effectively reflect the influence, caused by certain brand portfolio, on brand resources and corporate competence. In other words, this research can be used to be the guidance of brand structural optimization.