Research Of Cigarette Customer Knowledge Acquisition

Because of cigarette’s smoking feature, its raw material, formula, accessoriesand so on will have a greater impact on its quality, and affect the taste and quality,and then directly affect buyers’ favor degree. Sensory evaluation can examine andmonitor the quality objectively, and evaluators can grasp the overall quality. Thus, thesensory evaluation data’s role in developing new products should not beunderestimated. Therefore, the cigarette companies will regularly collect sensoryevaluation data in a certain period of time, in order to grasp the buyers’ favor degree,so that clear the direction of new products’ development. However, the sensoryevaluation data are numerous and dispersed, what’s worse, the statement expression isnot standardized, so there is a certain degree of difficulty in obtaining customerknowledge. Based on previous research foundation, this paper does some furtherresearch on obtaining customer knowledge, and the validity of the method is verifiedby experiment. Specific contents included as follows:(1) According to the characteristics of cigarette sensory data, this paper describeshow to deal with numerous evaluation data, and then obtain customer knowledge.This paper describes every step and technology we needed to achieve this goal indetail.(2) The consumer’s evaluation data will have a certain sentiment, and the waysof emotion reflecting are numerous, the main way here is by evaluation words. Inorder to compute the text sentiment and emotion intensity, some dictionaries areconstructed, including emotion dictionary, negative word dictionary and the degreeadverb dictionary.(3) Product feature extraction is the premise of analyzing customers’ sentimentalinclinations to specific product features. The actor of feature extraction is directly related to the accuracy of customer evaluation knowledge acquisition. Thus, based ofprevious studies, this paper proposes the extraction method of product feature wordsbased on Apriori algorithm.(4) In cigarettes evaluation data, each cigarette feature has its matchingevaluation word. In order to determine consumers’ sentiment on cigarette products,we need to make these two words matched. When matching the two words, we usedthe relation extraction method based on maximum entropy model.