DEVELOPMENT OF PERSONALIZATION SYSTEM OF SPECIALIZED WEB PORTAL
DOI:
https://doi.org/10.15588/1607-3274-2018-1-11Keywords:
personalization, JSM-method, neural network clustering, multi-agent system, informative portrait of user.Abstract
Context. The actual task of personifying a Web portal providing business services (telemedicine, consultations, remote monitoring,distance education, etc.) has been solved.
Objective - development of a personalization system for a web portal that provides specialized services, which allows to take into account
preferences of users for the improvement of quality of service, an acceleration of information search, an exception of uninteresting pages, and а customer retention.
Method. The generalized process personalization model of Internet service is offered. The method of adaptation of the Web-resource
based on the combination of agent and neural network technologies is proposed in a model which automatically generates content for certain
categories of Internet users. The document object model of site in a graph form to search of relevant information was proposed that allows the site personalization. The use of multi-agent structure allowed to realize interaction of the components of the developed model. The method includes the following actions: automatic generation of hypotheses, which determines the presence or absence of target properties of the user; analysis of the user’s behavior on his surfing the Internet that allows to give more relevant results; construction of information portrait for collection statistically significant set of information characteristics for the purpose of planning of further actions; parallel clustering of users with use of the self-organizing Kohonen maps for the purpose of an acceleration of processing big data. The self-organizing Kohonen maps are adapted to symmetric multiprocessing system for accelerating computations. Thus, the configuration of the computing system shall be a multiple of the dimension of the input data for reduction of computation time.
Results. For the proposed models and method, software and a web interface are developed. They are used to realization computing
experiments to verification of the models, valuation of the adequacy and study the properties of the model and method.
Conclusions. The conducted experiments have confirmed the proposed models and methods. The use of a set of methods and tools can
be used in practice to promote goods and services in the network, to provide various services or individual parts of it, for business development.
References
Xiao J., Zhang Y. Clustering of web users using session-based
similarity measures, Computer Networks and Mobile Computing,
Proceedings. 2001 International Conference on. IEEE,
, pp. 223-228. DOI: 10.1109/ICCNMC.2001.962600.
Chen L., Bhowmick S. S., Nejdl W. COWES: Web user clustering
based on evolutionary web sessions, Data & Knowledge
Engineering, 2009, Vol. 68, No. 10, pp. 867–885. DOI: 10.1016/
j.datak.2009.05.002.
Selvakumar K., Ramesh L. S., Kannan A. Enhanced K-Means
Clustering Algorithm for Evolving User Groups, Indian Journal
of Science and Technology, 2015, Vol. 8, No. 24, P. 1. DOI:
17485/ijst/2015/v8i24/80192.
Ganesan S., Sivaneri A. I. U., Selvaraju S. K. Evolving interest
based user groups using PSO algorithm, Recent Trends in
Information Technology (ICRTIT), 2014 International Conference
on, IEEE, 2014, pp. 1–6. DOI: 10.1109/ICRTIT.2014.6996196.
Andreeva K. A., Shajdurov R. S., Morgunov E. P. Primenenie
nejronnoj seti Kohonena dlja klassifikacii web-stranic
informacionno-poiskovoj sistemoj sajtov, Aktual’nye problemy
aviacii i kosmonavtiki, 2015, Vol. 1, No. 11, pp. 380–381.
Zerhari B., Lahcen A. A., Mouline S. Big data clustering: Algorithms
and challenge, Proc. of Int. Conf. on Big Data, Cloud and
Applications (BDCA’15), 2015.
Kurasova O. et al. Strategies for big data clustering, 2014 IEEE
th International Conference on Tools with Artificial Intelligence,
IEEE, 2014, pp. 740–747.
Axak N. Development of multi-agent system of neural network
diagnostics and remote monitoring of patient, Eastern-European
Journal of Enterprise Technologies, 2016, 4/9 (82), pp. 4–11.
Axak N., Korgut S., Komoda P. Decision support system for
intelligent site, Elektronika (LIV), No. 8/2013, pp. 52–59.
Anshakov O. M. DSM-metod: teoretiko-mnozhestvennoe
ob#jasnenie, NTI. Ser. 2. 2012, № 9.
Finn V. K. Induktivnye metody D.S. Millja v sistemah
iskusstvennogo intellekta. Chast’ I, Iskusstvennyj intellekt i
prinjatie reshenij, 2010, No. 3, pp. 3–21.
Shklovets A. V., Axak N. G. Visualization of high-dimensional
data using two-dimensional self-organizing piecewise-smooth
Kohonen maps, Optical Memory and Neural Networks, 2012,
Vol. 21, No. 4, pp. 227–232. DOI: 10.3103/
S1060992X12040066.
Downloads
How to Cite
Issue
Section
License
Copyright (c) 2018 N. G. Axak
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Creative Commons Licensing Notifications in the Copyright Notices
The journal allows the authors to hold the copyright without restrictions and to retain publishing rights without restrictions.
The journal allows readers to read, download, copy, distribute, print, search, or link to the full texts of its articles.
The journal allows to reuse and remixing of its content, in accordance with a Creative Commons license СС BY -SA.
Authors who publish with this journal agree to the following terms:
-
Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License CC BY-SA that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
-
Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
-
Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
