Method of data anomaly detection in the process of mobile applications installation

Tetiana Polhul, Andrii A. Yarovyi, Ryszard S. Romaniuk, Paweł Komada, Nursanat Askarova

2019

Closed access Metadata (CSL-JSON)
DOI 10.1117/12.2536855
Data confidence 85 high
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

In the given research the anomaly is considered as a fraud. The paper is devoted to the development of the method, aimed at detection of the data anomalies mobile applications installation. Anomaly (fraud) detection method on the basis of anomalies in data, which, unlike the existing ones, allows to solve the heterogeneous data resolution problem on which fraudulent decisions are made, reduce the size of the data and conduct a classification of users, is proposed. An experiment was conducted on the basis of a representative sample of 284,807 users acting from a real mobile application. The efficiency of the proposed method, the accuracy of which is 99.14%, is shown.