Font Size: a A A

SMS Seed Customers Mining Model And Algorithm Based On Content Analysis

Posted on:2017-08-30Degree:MasterType:Thesis
Country:ChinaCandidate:Z C HuangFull Text:PDF
GTID:2348330488477978Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
With the development of the Internet and the popularity of intelligent terminal,SMS(short message service), micro-blog, WeChat and other modern information transmission means are widely used, they have the advantages of convenient use,spread quickly. In contrast, SMS has low requirements about users' age, the level of knowledge and the function of the mobile terminal, that is the SMS's specific advantages, which is more suitable for broadcasting information, knowledge, family communication and other scenes. Seed SMS means carrying some specific information. SMS seed customers is the first round receiving/forwarding of the seed SMS, due to they have as high as possible interest and SMS forwarding quantity to specific seed SMS, so it has decisive influence for the transmission of information efficiency and cost. However, the current method of determining SMS seed customers is based on human's experience to find and filter, its efficiency and accuracy is not ideal. Therefore, it is great significance to control the spread of the seed SMS path,improve the efficiency of its spread and reduce transmission cost through mining the high quality SMS seed customers from the massive text message records.In order to mining SMS seed customers from massive text message records,control the spread of the seed SMS path and improve the efficiency of its spread, a SMS seed customers mining model and algorithm was proposed, which was based on content analysis. First of all, by analyzing the interest, randomness and one-way characteristics of customer forwarding messages, the tree model of customer forwarding messages were constructed. Secondly, by definition and application of comprehensive evaluation function to generate the optimal seed customers mining model, and based on the concept of close group to achieve SMS seed customers mining, and then using the real data of the telecom operators to carry on the empirical analysis, which verify the effectiveness of the model and the algorithm.Finally, this paper detailed analyzes the function requirements of telecom SMS enhancement application platform, and presents the system architecture design, thenimplements above research achievements, which is applied to the system.
Keywords/Search Tags:SMS seed customers, content analysis, mining model, mining algorithm, close group
PDF Full Text Request
Related items