Pattern Recognition with Fuzzy Objective Function Algorithms
Pattern Recognition with Fuzzy Objective Function Algorithms
A New Improved K-Means Algorithm with Penalized Term
GRC '07 Proceedings of the 2007 IEEE International Conference on Granular Computing
Agglomerative Fuzzy K-Means Clustering Algorithm with Selection of Number of Clusters
IEEE Transactions on Knowledge and Data Engineering
Clustering by competitive agglomeration
Pattern Recognition
Hi-index | 0.00 |
This research utilizes marketing research database the Taiwan telecom itself has together with Agglomerative Fuzzy K-Means to proceed fuzzy clustering analysis. The database content includes online behaviors and basic properties of clients, such as online motive, online frequency, salary, and gender. First, we use descriptive statistics to determine the difference in online behavior among different client clusters; these differences among clusters comprise indexes. Next, we compare the obtained indexes with experts' judgments to verify the precision of each index. These indexes can be used to estimate client's mobile online hours and the adaptive tariff plan. In addition, while approaching different cases, sales personnel can specifically query on significant questions within the index. Moreover, using these preidentification indexes, prolonged question analysis, especially on illogical answers, can be avoided. This can result in time saving and increase the number of cases handled, causing an overall improvement in industry performance.