摘要
在互联网背景下,针对传统竞争对手识别方法的局限,提出一种以用户评论为数据源的企业产品级竞争对手识别方法,旨在为企业的产品设计优化和竞争策略制定提供有力依据。首先,基于用户选择偏好界定候选竞争产品,并利用Python爬虫技术采集企业产品及候选竞争产品的在线评论;其次,运用Python分词技术结合频次统计与人工筛选,构建产品特征集与情感词集;再次,依托情感特征权重算法分析企业产品优劣势,形成特征优势与劣势集,构建产品向量空间模型并计算相似度;最后,识别出主要及次要竞争对手,为市场策略优化提供数据支撑。研究选取“高露洁”为实证分析案例,发现“高露洁”主要竞争对手为“佳洁士”“两面针”“冷酸灵”,优势相似且劣势程度相当;“牙博士”“黑人”因优势相似,在劣势特征上未达到相同水平,故列为次要竞争对手。
In view of the limitations of traditional competitor identification methods in the context of the Internet,this paper proposes a product-level competitor identification method based on user reviews as the data source,aiming to provide a strong basis for enterprises to optimize product design and formulate competitive strategies.Firstly,the candidate competitive products are defined based on the user's selection preference,and the online reviews of the company's products and the candidate competitive products are collected by using Python crawling technology.Secondly,the Python word segmentation technology is used to combine frequency statistics and manual screening to construct the product feature set and the sentiment word set.Thirdly,relying on the sentiment feature weight algorithm,the advantages and disadvantages of the company's products are analyzed,the feature strengths and weaknesses are formed,the product vector space model is constructed and the similarity is calculated.Finally,the main and secondary competitors are identified to provide data support for market strategy optimization.In this study,“Colgate” is selected as an empirical analysis case.The study finds that the main competitors of “Colgate” are “Crest” “Liangmianzhen” and “Lengsuanling”,with similar advantages comparable and disadvantages;“Dental Doctor” and “DARLIE” are listed as secondary competitors because they have similar advantages and do not reach a considerable level in disadvantage characteristics.
作者
范雨田
王晓慧
FAN Yutian;WANG Xiaohui(School of Management,Liaoning Normal University,Dalian 116082,China)
出处
《竞争情报》
2025年第2期22-28,共7页
Competitive Intelligence
关键词
在线评论
竞争对手识别
文本挖掘
情感分析
online reviews
competitor identification
text mining
sentiment analysis
