Keyword Extraction and Word Cloud Generation for Electricity Work Order Text Based on Attention Mechanism and TF-IDF Algorithm

With the digital transformation of the power industry, the accumulation of a large amount of work order text data provides a rich information resource for enterprises. However, how to effectively extract key information from these texts remains a challenge. In this paper, we propose a keyword extraction method for electric power work order text that combines the attention mechanism and the TF-IDF algorithm, and visualize the importance of the keywords using word cloud technology. The method first removes noise through text preprocessing, and then calculates the importance of each word using the TF-IDF algorithm, while using the attention mechanism to learn the correlation between words in order to further optimize the selection of keywords. The experimental results show that the proposed method significantly improves the accuracy of keyword extraction compared to the traditional TFIDF algorithm. In addition, the generated word cloud can clearly highlight the core topics in the text, which helps the electric power enterprises to quickly identify and respond to customer needs, and improve work efficiency and service quality. The research results in this paper can provide an effective method for text data analysis in the electric power industry.