Technical means is the important support of Weibo marketing

Micro Bo Marketing is flexible, need continuous interaction to gradually reveal the effectiveness of marketing, which seems to determine the micro-blog marketing can only be implemented by human intervention. In fact, it is the micro-blog through the interaction of the characteristics of value, the decision of micro-BO marketing can be effective technical means as the guide, the first to do data mining, on the basis of increased human intervention will be possible to achieve the maximum marketing effect.

The reality of marketing is the discovery, creation and delivery of value to meet the needs of the target users, in fact, the user as a central part of all actions to achieve value exchange http://www.aliyun.com/zixun/aggregation/8881.html. Micro-Blog created a large number of data and users, enterprises in the micro-blog even in the form of cluster account, it is also difficult to manually manage with all potential users at the same time, to discover all the relevant content of the enterprise. So the first thing microblogging marketing to solve is the enterprise and those user interaction, focus on those content.

One thing is for sure, users of Weibo users in a period of time role image is fixed, each user has its own characteristics and role positioning, regardless of attention, participation or forward evaluation, micro-blog interaction reflects the value of exchange, reflecting the social attributes between users. In other words, as long as users in micro-blog action, it will be able to locate the user's attributes, with the deepening of participation, the user's social attribute positioning can be more accurate and clear. Enterprises in micro-bonnet to find users, it is best to start from the role attribute. As long as can understand the user's attitude and will, can judge a user's role attribute, will take the attitude and the intention as the user Marketing Classification Foundation, its accuracy is higher than takes the user the rigid characteristic to classify. Attitudes and intentions are not easy to grasp, but the social network provides the possibility of analysis, users in the social network every action, behind their immediate attitude and willingness, the key is whether through technical means to judge, trace, Refine.

Now there is no perfect technical means to do this, micro-bonnet data mining is too complicated, not only in the user, the content of the mass, but also in the user characteristics are not obvious, unrelated to the fragmented content too much, the user behavior of the jump too strong. Do not say that it is a perfect data mining, even text mining is difficult to achieve.

Data mining is difficult to achieve in the short term, you can do information filtering first. The enterprise first according to own brand characteristic, the marketing goal and so on sets the related rule, uses the rule to represent the user model. Tracking the changes in micro-bonnet data, if the rules are met, the system will start to run, the system filtering part of the information filtered or preserved. The first is preprocessing: The system is required to preprocess the collected data, such as cleaning up "dirty" data, "short" data, too much jumping data, and data that is not consistent with user characteristics. The second is pattern discovery: using different mining algorithms to discover user behavior patterns and content classification. Furthermore, pattern Analysis: Choose meaningful patterns from the pattern collection of pattern discovery, and match users and content to extract the instantaneous dynamic characteristics of users. Refining the user and content into manual intervention, I believe that will be able to greatly improve the efficiency and effectiveness of micro-blog marketing (Shenlu Administration/text).