Information retrieval method and system based on multi-hop knowledge reasoning

A technology of information retrieval and knowledge reasoning, applied in the field of information retrieval, can solve problems such as insufficient utilization, unreliability, inability to reason and predict information, and achieve the effect of improving the value of use

CN114398895APending Publication Date: 2022-04-26NORTHEASTERN UNIV
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Publication Date
2022-04-26

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses an information retrieval method and system based on multi-hop knowledge reasoning. According to the method, a knowledge graph is constructed through information in a retrieval library, and a retrieval result is reasoned by utilizing the knowledge graph in an information retrieval process. The problem that data information in a retrieval library cannot be fully mined by a traditional information retrieval method is solved by means of a knowledge graph. By means of multi-hop knowledge reasoning, the property of prediction basis can be provided, information in an existing retrieval library is used for reasoning a result, and meanwhile the basis and source of the retrieval result can be provided for a user. The system designed for the method comprises a retrieval library information processing module, an information retrieval module, a knowledge graph learning module and an interactive interface. According to the method and the system, the information in the retrieval library is more fully retrieved, the retrieval content is reliable, and a retrieval basis is provided.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the technical field of information retrieval, in particular to an information retrieval method and system based on multi-hop knowledge reasoning. Background technique

[0002] Information Retrieval (Information Retrieval, IR) is a process of finding relevant information in the retrieval database according to the user's specific needs. With the advent of the era of big data, the scale of data continues to increase, and the types of data have become more diverse. How to perform accurate and efficient retrieval in massive data is an important issue. Traditional information retrieval methods, such as: Boolean logic retrieval, proximity retrieval, phrase retrieval, word truncation retrieval, field restriction retrieval, etc., are only simple matching, and have not fully explored the correlation between data in the context of big data.

[0003] The Knowledge Graph is a structured way to describe concepts, entities and their relatio...

Examples

Embodiment Construction

[0046] The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0047] Such as figure 1 As shown, the information retrieval method based on multi-hop knowledge reasoning in this embodiment is as follows.

[0048] Step 1: Establish a retrieval database and preprocess the information in the retrieval database;

[0049] The information in the retrieval library includes: database information and multimodal information. Among them, the preprocessing method of multimodal information is as follows:

[0050]S1: Obtain text data in multimodal information;

[0051] S2: Use known named entities to identify entities in the labeled text;

[0052] S3: Use the known relationship extraction method to obtain the relationship between entities;

[...