Explainable analytics: understanding causes, correcting errors, and achieving increasingly perfect accuracy from the nature of distinguishable patterns.

In addition to pursuing accurate analytics, it is invaluable to clarify how and why inaccuracy exists. We propose a transparent classification (TC) method. In training, data consist of positive and negative observations. To obtain positive patterns, we find the intersection between each of the two positive observations. The negative patterns are obtained in the same manner. Next, pure positive and pure negative patterns are established by selecting patterns that appear in only one type. In testing, such pure positive and pure negative patterns are used for scoring observations. Next, an observation is classified as positive if its positive score is not zero or if both its positive and negative scores are zero; otherwise, it is classified as negative. By experiment, TC can identify all positive (e.g., malignant) observations at low ratios of training to testing data, e.g., 1:9 using the Breast Cancer Wisconsin (Original) and 3:7 using the Contraceptive Method Choice. Without fine-tuned parameters and random selection, the uncertainty of the methodology is eliminated when using TC. TC can visualize causes, and therefore, prediction errors in a network are traceable and can be corrected. Furthermore, TC shows potential in identifying whether the ground truth is incorrect (e.g., identifying diagnostic errors).

The evaluation of trustworthiness to identify health insurance fraud in dentistry.

OBJECTIVE: According to the investigations of the U.S. Government Accountability Office (GAO), health insurance fraud has caused an enormous pecuniary loss in the U.S. In Taiwan, in dentistry the problem is getting worse if dentists (authorized entities) file fraudulent claims. Several methods have been developed to solve health insurance fraud; however, these methods are like a rule-based mechanism. Without exploring the behavior patterns, these methods are time-consuming and ineffective; in addition, they are inadequate for managing the fraudulent dentists. METHODS: Based on social network theory, we develop an evaluation approach to solve the problem of cross-dentist fraud. The trustworthiness score of a dentist is calculated based upon the amount and type of dental operations performed on the same patient and the same tooth by that dentist and other dentists. RESULTS: The simulation provides the following evidence. (1) This specific type of fraud can be identified effectively using our evaluation approach. (2) A retrospective study for the claims is also performed. (3) The proposed method is effective in identifying the fraudulent dentists. CONCLUSIONS: We provide a new direction for investigating the genuineness of claims data. If the insurer can detect fraudulent dentists using the traditional method and the proposed method simultaneously, the detection will be more transparent and ultimately reduce the losses caused by fraudulent claims.