Anomaly Detection Method For Aggregated Cellular Operator Data
| Autor: | Mark Bulygin, Dmitry Namiot |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: | |
| Zdroj: | Proceedings of the XXth Conference of Open Innovations Association FRUCT, Vol 28, Iss 1, Pp 41-47 (2021) |
| Druh dokumentu: | article |
| ISSN: | 2305-7254 2343-0737 |
| DOI: | 10.23919/FRUCT50888.2021.9347606 |
| Popis: | According to the research of the international agency ""We are social"" in April 2020, there are 5.16 billion mobile phone users in the world. This is 66% of the total population. This article is devoted to the search for anomalies in traffic data received from cellular operators. The first part of the article tells about what data is collected by cellular operators, how they are processed. A brief overview of methods for analyzing these data is provided. Further, the structure of the data that is collected from cellular operators by a special department of the city of Moscow is presented. In the second part of the article, methods for anomaly detection are analyzed, taking into account the specifics of the available data, and the advantages and disadvantages of its use are revealed. Then, a proprietary method for anomaly detection is proposed, which is based on the properties of traffic data. The work of the proposed method is demonstrated in several computational experiments. Further directions of work in the field of aggregated data analysis of cellular operators are proposed. |
| Databáze: | Directory of Open Access Journals |
| Externí odkaz: |
|