洪健哲
洪健哲
社群媒體為基之市場區隔趨勢預測方法與技術研發
Development of Method and Technology for Social Media-based Market Segment Trend Prediction
社群媒體、市場區隔預測、消費者影響力、市場趨勢分析
social media、market segment forecasting、consumer influence、market trend analysis
「市場」為供給者與需求者進行商品或勞務交易的地方，「行銷」(Marketing) 則是重要的商務活動之一。了解市場區隔、選擇目標市場、提供適合目標市場之商品，並施以適當的行銷策略，為當前行銷策略之核心。傳統上，企業都以市場調查法分析市場現況，除了常發生花費過多的人力、時間與成本，也存在無法及時與正確反映市場現況以及變化趨勢等問題。
隨著數位時代到來，社群媒體已成為人們重要的溝通與資訊分享工具，也是企業分析市場之重要資料來源。當前針對社群媒體為對象之市場分析相關研究，多以目標市場整體變化為主，缺乏對各市場區隔消長以及市場內元素對市場區隔消長的影響分析。因此，如何利用社群媒體資料預測市場區隔消長變化趨勢，供企業及時掌握市場變化以提升競爭優勢，為一重要研究課題。
針對市場區隔趨勢預測之需求、社群媒體內容之可用性，本研究之主要目的在設計一社群媒體為基之市場區隔趨勢預測方法，並開發其實現技術。首先從系統與微觀的角度，分析市場內元素間之影響力，設計市場影響力模型。再依此模型，設計針對社群媒體內容之市場影響力分析方法。接著設計以影響力為基之市場區隔趨勢預測方法並開發實現技術。為驗證方法之有效性，本研究首先經由實驗，取得較適當之預測參數，再依此參數進行市場區隔變化預測與比較。以實際數值與預測之結果計算兩者間的誤差，得出本研究所提之影響力為基的區隔趨勢預測方法，誤差平均值為0.0936，相較於僅利用區隔人數進行區隔趨勢預測方法的誤差平均值0.0971，約降低3.7%。故以影響力為基的市場區隔趨勢預測是可行且有效的。
Understanding the market segmentation, selecting the target market, providing products that fit the target market, and applying appropriate marketing strategies are at the heart of current marketing strategies. Traditionally, companies have used the market research method to analyze the current market conditions. In addition to the excessive manpower, time and cost, there are also problems such as the inability to timely and correctly reflect the current market conditions and trends.
With the advent of the digital era, social media has become an important communication and information sharing tool for people, and an important source of information for the analysis of the market. At present, the market-related research on social media is mainly based on the overall change of the target market. There is a lack of analysis on the impact of market segmentation and the influence of market elements on market segmentation. Therefore, how to use social media data to predict the trend of market segmentation and change, for enterprises to grasp market changes in time to enhance competitive advantage, is an important research topic.
The main purpose of this study is to design a social media-based market segment trend forecasting method and develop its implementation technology. The results show that the average error value of the influence-based segment trend prediction method of this study is 0.0936, which is about 3.7% lower than that of the 0.0971 method using the segmentation trend prediction method. Therefore, market segmentation trend prediction based on influence is feasible and effective.