Method and system for performing multi-dimensional combined retrieval on enterprise information
An enterprise information, multi-dimensional technology, applied in the field of multi-dimensional combined retrieval of enterprise information, methods and systems, can solve problems such as time-consuming, labor-intensive, inability to effectively locate, and inability to meet user needs, to ensure accuracy, The effect of reducing bounce rate and improving experience
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0036] A method for multi-dimensional combined retrieval of enterprise information, specifically comprising the following steps:
[0037] Step 1. Basic construction: build a fastapi+python microservice framework, collect enterprise information on the Internet through the docker engine, and construct a dictionary of the collected enterprise information through NLP related technologies to generate an initial search tag library corresponding to enterprise information;
[0038] Step 2. Improve learning: import the initial search tag library corresponding to the enterprise information generated in step 1 into the bert model, input the enterprise information sentence into the bilstm+crf model architecture, and compare the search tags stored in the initial search tag library according to the output results Carry out fine-tuning, after perfecting the dictionary constructed in step 1, extend the adjusted search tags to the initial search tag library in step 1 to form a fine-tuned search...
Embodiment 2
[0042] This embodiment, as an improvement of the previous embodiment, is a method for multi-dimensional combined retrieval of enterprise information, which specifically includes the following steps:
[0043] Step 1. Basic construction: Build the microservice framework of fastapi+python, collect enterprise information on the Internet through the docker engine, and construct a dictionary of the collected enterprise information through NLP related technologies. It should be noted that NLP related technologies specifically include Text retrieval, for large-scale data retrieval; machine translation, for cross-language translation; text classification; information extraction, for extracting desired information from irregular text; Words / phrases are marked with corresponding tags; text summaries are used to focus on the most core parts from a given text and automatically generate summaries; generate an initial search tag library corresponding to enterprise information. It should be no...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 



