一句话：提出 APPFIRE，结合 NER、BERT 与主题建模，从用户描述中自动挖掘并推荐相似移动应用特性。
移动应用开发需要提取领域特定、必要且创新的功能特性，以契合用户需求与市场动态。手动识别这些特性对产品经理而言成本高且困难。本文提出 APPFIRE，自动根据用户提供的应用描述找到相似应用，并挖掘与推荐应用特性。方法上，它先用命名实体识别（NER）抽取特性，再用 BERT（以及 SBERT 嵌入）结合主题建模识别相似应用；其中采用非负矩阵分解（NMF）做主题建模的模型表现最佳，F1 达到 87.38%。评估表明该数据驱动流程能有效支持应用特性发现与推荐。
Mobile app development necessitates the extraction of domain specific, essential and innovative features, aligning with user needs and market dynamics. Identifying features to provide competitive edge to the app developers, is a non-trivial task that is often performed manually by product managers. This study addresses the challenge of mining and recommending app features by automatically identifying similar apps corresponding to the description of apps provided by the user. The proposed approach, APPFIRE, integrates Named Entity Recognition (NER) for feature extraction and BERT (Bidirectional Encoder Representations from Transformers) coupled with Topic Modeling for identifying similar apps. Our top-performing model, utilizing Non-negative Matrix Factorization (NMF) for Topic Modeling with SBERT embeddings, achieves an F1 score of 87.38%.