Please enable JavaScript.
Coggle requires JavaScript to display documents.
Age Prediction version 0
Given CALL, SMS (including text) and browsing…
Age Prediction version 0
- Given CALL, SMS (including text) and browsing data, predict age of users
References
Findings
- Postpaid thường có thông tin demographics trong khi prepaid là anonymous
- D. Nguyen, R. Gravel, D. Trieschnigg, and T. Meder, “‘TweetGenie: automatic age prediction from tweets’ 6, Sep. 2013.
- Argamon, S., Koppel, M., Pennebaker, J. W., & Schler, J. (2009). Automatically Profiling the Auth
or of an Anonymous Text.
Communications of the
https://doi.org/10.1145/1461928.1461959
- Kosinski, M., Stillwell, D.,
&Graepel, T. (2013). Private Traits and Attributes are Predictable from Digital Records of Human Behavior.
http://doi.org/10.1073/pnas.1218772110
- Marquardt, J., Farnadi, G., Vasudevan, G., Moens, M. F., Davalos, S., Teredesai, A., & De Cock, M. (2014). Age and Gender Identification in Social Media.
- Mechti, S., Jaoua, M., & Belguith, L. H. (2014). Machine Learning for Classifying Authors of Anonymous Tweets, Blogs, Reviews and Social Media
- Nguyen, D., Smith, N., &
Rosé, C. (2011). Author Age Prediction from
Text using Linear Regression.
- Nguyen, T., Phung, D., Adams, B., & Venkatesh, S. (2011). Prediction of age, sentiment, and connectivity from social media text.
- Sap, Maarten, Eichstaedt, Johannes, Kern, Margaret L., Stillwell, David, Kosinski, Michal, Ungar, Lyle H., & Schwartz, H. Andrew. (2014). Developing Age and Gender Predictive Lexica over Social Media
- L. Sloan, J. Morgan, P. Burnap, and M. Williams, “Who Tweets? Deriving the Demographic Characteristics of Age, Occupation and Social Class from Twitter User Meta Data,”
- Predict age based on GPS
- Inferring User Demographics and Social Strategies in Mobile Social Network
- Demographic Prediction Based on User’s Browsing Behavior
-
-
Assumptions
Network characteristics
- Young people are very active in broadening their social circles, while seniors tend to keep close but more stable connections.
- Female put more attention on cross-generation interactions than male.
- Upon entering into middle-age, people’s attention to opposite gender triads quickly disappears. However, the insistence and social investment on same-gender social groups (‘FFF’ and ‘MMM’) lasts for a lifetime.
EDA
Input data and features
Labels
202,533 obs
Trong 1 tháng từ 01/04 đổ lại thì chỉ lấy ra đc 136,834 labeled obs
-
-
user_profile
export 58 user_profile features for 136,834 labels
-
-
Training input = join of call, sms, num socket, user profiles => 62,376 obs
-
Features
-
-
-
-
SMS Vocab
151 features
-
hành vi: học, cơm, đón, yêu, ...
-
-
emoticon: yahoo, facebook
-