企業藉由行銷活動吸引新顧客與留住舊顧客，而產品定位分析則是企業擬定行銷策略的方法之一，以得知消費者對產品特徵之重視度，使企業能夠了解市場需求。現下，隨著各式網路媒體的蓬勃發展，消費者無不透過社群網站撰寫評論，分享其意見及心得，這些文章透露出消費者對於產品的重視度及功能需求。因此，本研究試圖以馬斯洛需求層級理論（Maslow's hierarchy of needs theory）為基，發展一個產品定位分析（Product positioning）機制，利用意見探勘（Opinion mining）、凝聚式階層分群（Agglomerative hierarchical clustering, AHC）及倒傳遞類神經網路（Back-propagation neural network, BPNN）等方法，分析消費者對產品之評論，探勘客群對特定產品所重視的產品特徵及各特徵對應至馬斯洛理論之層級，以建立一消費者的產品馬斯洛模型。此機制將可提供企業審視產品在目標市場之正確定位，並作為新產品開發改良的參考。 Through marketing activities, enterprises can attract new customers and retain valuable customers. Product positioning analysis is one of the important steps to develop an effective marketing strategy, which aims to find out the degree of attention of consumers for particular features of products. Understanding product positioning in customer minds will enable enterprises to research and develop the novel products that can completely match consumer demands. Currently, with the prosperous development of various network social media platforms, on which consumers can share their reviews about products. These reviews often tend to reveal the degree of recognition and necessity for products from consumers. This study tries to design a product positioning analysis mechanism based on Maslow's hierarchy of needs theory, so that enterprises can understand the level of demand in target audiences psychological dimension in low-cost and automated way. This study designs a product's Maslow model that is used to analyze what product features TA care about, by means of machine learning methods including opinion mining, agglomerative hierarchical clustering (AHC), and back-propagation neural network (BPNN) to mining consumers' opinion on social media platforms and then map each features to the five levels of Maslow's hierarchy of needs. The mechanism will provide information of product positioning in target market to enterprises, so as to support refining product positioning strategies, improving features of products, and making decisions.
