Hello everyone, I am Weixue AI. Today, I will introduce to you the practical application of knowledge graph 12-recipe field intelligent question answering system, which realizes recipe question and answer. This project is based on py2neo and neo4j graph database, and applies knowledge graph to the recipe field. By building a recipe knowledge graph, a simple question answering system for recipe ingredients is realized. Users can quickly obtain simple recipe ingredient information through the question-and-answer system.
1. Recipe Knowledge Graph Technology Selection
1.py2neo and neo4j: In addition to py2neo, there are other python libraries that can connect to neo4j graph databases, etc.
2. Database tools: You can use other data processing libraries, such as NumPy, SciPy, scikit-learn, etc. In addition, other data formats can also be used for storage and import, such as JSON, RDF, etc.
3. Natural language processing library: In addition to jieba and nltk, there are other Chinese text processing libraries, such as HanLP, THULAC, etc. In terms of syntax analysis, you can use open source syntax analyzers, such as Stanford Parser, Berkeley Parser, etc. In terms of entity recognition, deep learning models such as BERT and CRF can be used.
4. Question answering system framework: In addition to Flask, there are other web application frameworks, such as Django, Tornado, FastAPI, etc. When developing a question answering system, other natural language processing frameworks such as Rasa, SpaCy, etc. can also be used.
Recipe knowledge map visualization interface:
2. Sample data
We selected a few simple recipes as sample data, the data format is csv, which contains information such as dish names, ingredients, seasonings and steps.
Sample data: </