Internet information collection and association method

Abstract

The invention discloses an Internet information collection and association method. The method comprises the steps that (1) an index system is constructed according to Internet information of a certain individual; (2) a target set O is solved according to a funnel model; (3) a target set O is solved according to association rules; (4) a final target set, namely O=OUO is acquired; and (5) the target set O=OUO is output. According to the method, the funnel model is adopted, the influence of traditional subjective assignment weight is eliminated, thus, the target set is most approximate to the objective fact, and the system can well grab the target set which is needed; the association rules and the apriori algorithm are also adopted, under the condition that the degree of support and confidence of the model is guaranteed, logic is strict, irrelevant sets are eliminated, all target sets are reserved, and therefore the target set meeting the intention is found; the method of using the funnel model and the apriori algorithm to screen data has advantages which are incomparable to a single rule, the target set is better approximate to expectation, and the precision of Internet information collection and association is remarkably improved.

Description

technical field [0001] The invention relates to a method for processing Internet information, in particular to a method for collecting and associating Internet information. Background technique [0002] With the rapid development of the Internet and the continuous development of data collection technology, the era of big data has already arrived. However, with the rapid development of the Internet, the large increase in the amount of information on the Internet has made it impossible for users to obtain the part of information that is really useful to them when faced with a large amount of information, and the efficiency of using information is reduced instead. This is the so-called The "information overload" (informationoverload) problem. How to help users quickly find valuable information in massive amounts of data, and make dark information in the network available to users has become an urgent problem to be solved. [0003] At present, the recommendation system is a to...

AI Technical Summary

Problems solved by technology

[0004] The user profile of the traditional recommendation system collects data through web crawlers, first preprocesses the data, and then screens out consumers through weight and attenuation factors. The distribution of weight factors is fuzzy and subjective, resulting in incomplete screening results. Satisfactory; when there are many attribute values, the assignment of weights becomes more difficult, resulting in the results screened out by the algorithm not meeting the purpose of precision marketing, that is, no potential consumers can be found or the proportion of potential consumers in the results very low

Images