Article

Article title MODELING THE PREDICTION OF THE USER RETENTION METRICS BASED ON NEURAL NETWORKS
Authors M. V. Sychugov
Section SECTION III. EVOLUTIONARY MODELING AND BIOINSPIRED ALGORITHMS
Month, Year 04, 2018 @en
Index UDC 004.032.26
DOI
Abstract This article is devoted to solving the problems of modelling the prediction of retention metrics in the field of mobile development, where you need to have data at what time users leave the application for the purpose of taking measures to increase the lifetime and level of monetization indicators. Since in most applications, significant user churn happens in the initial time interval of use, so the timely solution of problems in the application is useful both in commercial terms and for the study of user behavior, which in turn will allow developing a more cost-effective product. The aim of the work is to develop a model for predicting metrics that are fairly close to real indicators. The proposed model based on a neural network trained in the back-propagation method provides the ability to predict different time periods relative to the number of remaining users in the application. The backpropagation method is chosen because of its sufficient efficiency and highly adaptable architecture with an arbitrary number of layers, inputs and outputs. Scientific novelty is represented by combining neural networks with heuristics based on key data categories. These data represent the segmentation of mobile users into various categories, which allow them to capture aspects of their behavior for a more adaptable perdiction of retention of different target groups. The results of the prediction are given for different time intervals typical for the given mobile application use.

Download PDF

Keywords Neural networks; method of back propagation; retention of mobile users; predicting systems; time series.
References 1. Hadiji F., Sifa R., Drachen A., Thurau C., Kersting K., and Bauckhage C. Predicting Player Churn in the Wild, in Proc. of IEEE CIG, 2014.
2. Sifa R., Hadiji F., Runge J., Drachen A., Kersting K., and Bauckhage C. Predicting Purchase Decisions in Mobile Free-to-Play Games, in Proc. of AAAI AIIDE, 2015.
3. Kravchenko Yu.A. Tekhnologiya analiza nadezhnosti adaptivnykh informatsionnykh sred [Technology reliability analysis of adaptive information environments], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2010, No. 12 (113), pp. 103-108.
4. Smith L.S., Gheyas I.A. A Neural Network Approach to Time Series Forecasting, Proceedings of the World Congress on Engineering, London, 2009.
5. Osovskiy S. Neyronnye seti dlya obrabotki informatsii [Neural networks for information processing]: transl. from polish by. I.D. Rudinskogo. Moscow: Finansy i statistika, 2004, 344 p.
6. Bova V.V., Kravchenko Y.A., Kureichik V.V. Development of distributed information systems: ontological approach, Advances in Intelligent Systems and Computing, 2015, Vol. 349,
pp. 113-122.
7. Kravchenko Yu.A. Otsenka kognitivnoy aktivnosti pol'zovatelya v sistemakh podderzhki prinyatiya resheniy [Evaluation of cognitive activity of the user in decision support systems], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2009, No. 4 (93), pp. 113-117.
8. Kravchenko Yu.A. Tekhnologiya analiza nadezhnosti adaptivnykh informatsionnykh sred [Technology reliability analysis of adaptive information environments], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2010, No. 12 (113), pp. 103-108.
9. Kuliev E.V., Lezhebokov A.A., Kravchenko Yu.A. Roevoy algoritm poiskovoy optimizatsii na osnove modelirovaniya povedeniya letuchikh myshey [Swarm algorithm search engine optimization is based on modeling the behavior of bats], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2016, No. 7 (180), pp. 53-62.
10. Chernyshova G.Yu. Intellektual'nyy analiz dannykh: ucheb. posobie dlya studentov spetsial'nosti 080801.65 «Prikladnaya informatika (v ekonomike)» [Data mining: textbook for students of specialty 080801.65 " Applied Informatics (in Economics)"]. Saratov: Saratovskiy gosudarstvennyy sotsial'no-ekonomicheskiy universitet, 2012, 92 p.
11. Zaytsev A.A., Kureychik V.V., Polupanov A.A. Obzor evolyutsionnykh metodov optimizatsii na osnove roevogo intellekta [Overview of evolutionary optimization techniques based on swarm intelligence], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2010, No. 12 (113), pp. 7-12.
12. Kureychik V.V., Kureychik Vl.Vl. Bioispirirovannyy poisk pri proektirovanii i upravlenii [Biospherology search in the design and management], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2012, No. 11 (136), pp. 178-183.
13. Gladkov L.A., Kureychik V.M., Kureychik V.V. Geneticheskie algoritmy [Genetic algorithm]. Moscow: Fizmatlit, 2006, 320 p.
14. Rodzin S.I., Kureychik V.V. Sostoyanie, problemy i perspektivy razvitiya bioevristik [State, problems and prospects of development of bio-heuristics], Programmnye sistemy i vychislitel'nye metody [Software systems and computational methods], 2016, No. 2, pp. 158-172.
15. Rodzin S.I., Kureychik V.V. Teoreticheskie voprosy i sovremennye problemy razvitiya kognitivnykh bioinspirirovannykh algoritmov optimizatsii [Theoretical questions and contemporary problems of the development of cognitive bio-inspired algorithms for optimization], Kibernetika i programmirovanie [Cybernetics and programming], 2017, No. 3, pp. 51-79.
16. Kureychik V.V., Rodzin S.I. O pravilakh predstavleniya resheniy v evolyutsionnykh algoritmakh [About the rules for the submission of solutions in evolutionary algorithms], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2010, No. 7 (108), pp. 13-21.
17. Zaporozhets D.Yu., Kudaev A.Yu., Lezhebokov A.A. Mnogourovnevyy algoritm resheniya zadachi parametricheskoy optimizatsii na osnove bioinspirirovannykh evristik [A multilevel algorithm for solving the problem of parametric optimization based on bio-inspired heuristics], Izvestiya Kabardino-Balkarskogo nauchnogo tsentra RAN [zvestiya of Kabardino-Balkar scientific center of RAS], 2013, No. 4 (54), pp. 21-28.
18. Kureychik V.M. Osobennosti postroeniya sistem podderzhki prinyatiya resheniy [Features of construction of systems of support of acceptance of decisions], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2012, No. 7 (132), pp. 92-98.
19. Kureychik V.M., Kazharov A.A. Ispol'zovanie shablonnykh resheniy v murav'inykh algoritmakh [The use of standard solutions in ant colony optimization algorithms], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2013, No. 7 (144),
pp. 11-17.
20. Gladkov, L.A., Kravchenko Y.A., Kureichik V.V. Evolutionary Algorithm for Extremal Subsets Comprehension in Graphs, World Applied Sciences Journal, 2013, Vol. 27 (9), pp. 1212-1217.

Comments are closed.