Design And Implementation Of "Internet +" Agricultural Town Resource Management And Service System

With the rapid growth of information on the Internet today,logistics and e-commerce have developed rapidly in the field of agriculture.People hope that they can quickly get the information they want from the Internet.Combining the characteristics of timeliness and multiple information categories in the agricultural sector,this paper combines the agricultural information service system,text classification system,and information push system to discuss the realization of intelligent push services for agricultural thematic information.First,design the overall framework and flow of the system.Research the use of web crawlers to collect cultural information related to agriculture on the web and build search engines for agricultural topics,so that users can accurately find the needed agricultural information.In the process of using the search engine by the user,a user interest model is established according to the user's search history and browsing history,so that the system can analyze which kind of information the user is interested in.In the information push service,a text classifier is used to find information that the user is interested in and push.An extended Heritrix web crawler collects text information related to agriculture on the web and classifies and stores it using a text classifier,then uses the Lucene full-text indexing framework to build an indexing library for these texts,and establishes a search engine system for agricultural topics to improve the general search engine.Search-oriented information is not deep enough to search for information,improving the accuracy of information retrieval.On the issue of classification of agricultural text information,a method of classifying agricultural information texts collected by web crawlers using a text classification classifier is first defined.The problem of classification errors caused by the uneven distribution density of samples at the category boundaries of the KNN text classifier is discussed.An improved KNN classifier based on K-means clustering is proposed.The clustering classifier is used as the first classifier and the samples in the cluster are used as the classifier.The experimental results show that the improved scheme is effective.Sex.When the user interest model is established,the user's interest in the keyword is calculated using the keyword access frequency and the average access time.Considering that the user's focus is changed,the interest attenuation factor is introduced to eliminate the keywords that the user has not noticed for a long time,and the user interest model is updated.Then the query expansion scheme based on user interest model is discussed.When designing the information push service scheme,the interest degree of the keyword in the user interest model is used as a weight,and an improved text classifier is used to find the information that the user is most interested in and pushed to the user.Finally,the development of the system Web server,taking into account the current popularity of smart phones,mobile phone mobile app is also designed to implement the information interface and information push between the system and the user.Then the system was tested for functional modules.