Clustering Internet Usage Behaviours with SOM Neural Networks


Celenk U., Ucan O.

World Congress on Engineering and Computer Science, San-Francisco, Kostarika, 24 - 26 Ekim 2012, ss.967-972 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası:
  • Basıldığı Şehir: San-Francisco
  • Basıldığı Ülke: Kostarika
  • Sayfa Sayıları: ss.967-972
  • İstanbul Üniversitesi Adresli: Evet

Özet

According to different needs of users, there are different consumption habits. Consumption habits of people, which have the same age group or the same professions, are similar. A type of internet usage habits of people in this way is one of these habits. In recent years, developments in technology, GSM, and especially with 4G mobile internet usage have found applications in many areas of daily life. Enter to internet, wherever users need to, creates freedom. Messaging, media, finance and many different needs can be met through this connection. Users' occupation, age, gender, location, usage patterns according to different characteristics such as income level and the relevant properties are similar to each other according to the amount of internet usage (in Mb Download) connected to internet and internet usage frequency and duration of exposure can be clustered. SOM type of study, personal internet usage by artificial neural networks (data of the CDR) process and their profession, age, gender, location is to cluster usage patterns according to the values.