Case retrieval network top-k extension activation method based on virtual case nodes
A top-k, case technology, applied in the field of top-k extension activation of case retrieval network based on virtual case nodes, can solve the problems of exponential decline in efficiency, affect retrieval efficiency, uselessness, etc., and achieve the effect of improving retrieval speed.
Inactive Publication Date: 2017-03-15
NANCHANG UNIV
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, there are a large number of different types of information entities in a large-scale CNR. Due to the small-world phenomenon (Kleinberg, 2000), any two information entities of the same type may have similar correlations. At this time, the activation of a large number of information entities will Affects retrieval efficiency and returns a large number of useless case results
Although the retrieval method of CRN can find the optimal solution, when the amount of data continues to increase, its efficiency drops exponentially.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0054] The technical solutions of the present invention will be further specifically described below through the embodiments and in conjunction with the accompanying drawings.
[0055] Firstly, the theoretical basis used in the present invention is introduced.
[0056] A graph integrating case representation and retrieval in the case retrieval process of Case Retrieval Nets (CRN) (Lenz, 1999, Lenz et al., 1996). Before retrieval, first compile the case base into a CRN, then obtain the information entity E' in the query, select the initial node in the CRN through E', and then expand and spread in the CRN through the activation function until the activation termination condition is met, and finally the The activated nodes are sorted and output as the retrieval result. The advantage of this kind of network is that it adopts a modular data structure, and the retrieval is efficient. Its efficiency is only related to the number of activated information entities, and the absence of ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a case retrieval network top-k extension activation method based on virtual case nodes. The method includes 1 retrieving the initial information entity in the network according to activation case query;2 activating case nodes relevant to the information entity, calculating similarity values of case nodes and query, and adding to the top-k retrieval results in the descending order of similarity values; 3 determining whether activation is further performed or not according to the virtual case nodes, if so, repeating step 2, if not, terminating the activation and returning to the top-k retrieval results. The virtual case nodes are relevant to one hypothesis and one following information entity to be activated, and are not allowed to exist in the case retrieval network actually. According to the method, by controlling the extension activation process of the case retrieval network through virtual case nodes, the problem that large majority of information entity nodes are activated during top-k query can be avoided effectively, accordingly the number of extension is reduced, and the retrieving speed is increased.
Description
technical field [0001] The invention relates to the technical field of case reasoning, in particular to a technical scheme of a virtual case node-based top-k expansion and activation method for a case retrieval network. Background technique [0002] Case-based reasoning (CBR for short) is an important reasoning technology emerging in the field of artificial intelligence in recent years. When people solve problems, they usually recall the experience and processing process of similar problems in the past, and make appropriate modifications to these experiences and processing processes in the environment of new problems, and reuse them to solve new problems. CBR is an artificial intelligence method that simulates this problem-solving method of people. These processes are reflected in CBR as the CBR processing model. The 4R model proposed by Aamodt is the most classic (Aamodt et al., 1994). The core issues include case representation, case similarity calculation, case retrieval...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/30
CPCG06F16/24564G06F16/2462
Inventor 李铭
Owner NANCHANG UNIV