Outline

  • Abstract
  • Keywords
  • 1. Introduction
  • 2. Related Work
  • 3. Communication Behavior Indicator and Fcm Algorithm
  • 3.1. Communication Behavior Indicator
  • 3.2. User Segmentation with Fuzzy -Means Clustering
  • 4. User Behavior Patterns
  • 4.1. Communication Behavior Clustering
  • 4.2. Communication Behavior and Mobility
  • 4.3. Group Clustering
  • 4.4. Results and Discussions
  • 5. Conclusions and Future Work
  • Acknowledgments
  • References

رئوس مطالب

  • چکیده
  • 1. مقدمه
  • 2. کارهای مرتبط
  • 3. شاخص رفتار ارتباطات و الگوریتم FCM
  • 3.1. شاخص رفتار ارتباطاتی
  • 3.2. تقسیم بندی کاربر با خوشه بندی c-means فازی
  • 4. الگوهای رفتاری کاربر
  • 4.1 خوشه بندی رفتار ارتباطات
  • 4.2 رفتار ارتباطی و تحرک
  • 4.3. خوشه بندی گروه
  • 4.4. نتایج و بحث
  • 5. نتیجه گیری و کارهای آتی

Abstract

Understanding the intelligence of human behaviors by mining petabytes of network data represents the tendency in social behaviors research and shows great significance on Internet application designing and service expansion. Meanwhile, the running mobile networks that generate huge data can be the best social sensor for these studies. This paper investigates a practical case of mobile network aided social sensing which uncovers some features of users’ behaviors in mobile networks by intelligently processing the big data. The paper studies the users’ behaviors with regard to communication, movement, and consumption based on large user data sets. The main contribution of the study is some findings on the relations among these behavior features. We find that the users’ calling behaviors are different despite their monthly expenditures being similar, though different consumption level users may have similar communication behaviors. We also find that statistically users with the higher mobility contribute more ARPU than those with lower mobility. Additionally, we also find that the top consumption level users are the most “lonely” ones by exploring the movement clustering patterns of users. These findings are significant to instruct marketing strategies for telecommunication industry.

Keywords: - - -

5. Conclusions and future work

Due to the popularity of mobile phone usage and the fast development of computer technology, we can study human behaviors based on telecommunication records. Comparing with questionnaires, the telecommunication records can accurately reflect people’s real life, thus the data analysis based on these records is more convinced. Our work described in this paper is significant to instruct ISP CRM and help anthropology study. We proposed an algorithm of user segmentation based on FCM. Using this algorithm, we divide mobile users into 3 clusters considering their communication behaviors. Integrating with the traditional pyramidal model based on ARPU, we got in-depth behavior patterns of mobile users.

However, limited by the supporting projects purpose and privacy concerns, we can only use the 16-week long and a little bit outdated data. Nevertheless, the proposed methods in the paper have already been submitted to the service provider and used in their system. According to their feedback, the methods work fine for other later CDR data. Therefore, we believe the 16-week long data is enough for the research itself and suitable for ISP’s CRM purpose.

For future work, we will work on the technologies for privacy preservation in order to appropriately make use of user data for better mobile services. And we hope to model users’ long term behaviors and the behavior dynamics with the support of data of longer period, which can be even more valuable to the service providers.

دانلود ترجمه تخصصی این مقاله دانلود رایگان فایل pdf انگلیسی