Network Community Information Resource Optimization Organization Based on Keyword Co-occurrence Relationship in Meishitianxia Website

The Meishitianxia website is a Chinese network platform focusing on the field of food, which provides a communication platform integrating cooking demonstration, food exchange and healthy diet knowledge sharing for the majority of food lovers. At present, the website provides users with a relatively simple navigation system, including menu classification of cuisines and other categories. This navigation mode can only enable users to obtain unstructured information in a major category, and does not reflect the correlation between the keywords in the categories, making it difficult for users to locate the required information clearly and quickly. Taking the Cantonese cuisine section of the Meishitianxia website as an example, keywords from existing culinary encyclopedias and other food websites are extracted to form a custom keyword table that includes eight categories: Dish Categories, Suitable Crowd, Flavor, Cooking Technique, Difficulty, Time Consumption, Kitchen Utensils, and Ingredients. Use a Python webscraper to collect user interaction content from the website, and employ the custom keyword table as a custom dictionary for the jieba library to segment the collected website text content. The Python Count function is utilized for conducting keyword frequency analysis. Employ the NumPy library in Python to generate a co-occurrence matrix of keywords. Calculate the co-occurrence intensity between keywords using the Ochiia coefficient. Based on the keyword frequency and the co-occurrence intensity between keywords, construct a multi-level classification navigation system for the Cantonese cuisine section of the Meishitianxia website.