TY - GEN
T1 - App2Vec
T2 - 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016
AU - Ma, Qiang
AU - Muthukrishnan, S.
AU - Simpson, Wil
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/11/21
Y1 - 2016/11/21
N2 - We design a way to model apps as vectors, inspired by the recent deep learning approach to vectorization of words called word2vec. Our method relies on how users use apps. In particular, we visualize the time series of how each user uses mobile apps as a 'document', and apply the recent word2vec modeling on these documents, but the novelty is that the training context is carefully weighted by the time interval between the usage of successive apps. This gives us the app2vec vectorization of apps. We apply this to industrial scale data from Yahoo! and (a) show examples that app2vec captures semantic relationships between apps, much as word2vec does with words, (b) show using Yahoo!'s extensive human evaluation system that 82% of the retrieved top similar apps are semantically relevant, achieving 37% lift over bag-of-word approach and 140% lift over matrix factorization approach to vectorizing apps, and (c) finally, we use app2vec to predict app-install conversion and improve ad conversion prediction accuracy by almost 5%. This is the first industry scale design, training and use of app vectorization.
AB - We design a way to model apps as vectors, inspired by the recent deep learning approach to vectorization of words called word2vec. Our method relies on how users use apps. In particular, we visualize the time series of how each user uses mobile apps as a 'document', and apply the recent word2vec modeling on these documents, but the novelty is that the training context is carefully weighted by the time interval between the usage of successive apps. This gives us the app2vec vectorization of apps. We apply this to industrial scale data from Yahoo! and (a) show examples that app2vec captures semantic relationships between apps, much as word2vec does with words, (b) show using Yahoo!'s extensive human evaluation system that 82% of the retrieved top similar apps are semantically relevant, achieving 37% lift over bag-of-word approach and 140% lift over matrix factorization approach to vectorizing apps, and (c) finally, we use app2vec to predict app-install conversion and improve ad conversion prediction accuracy by almost 5%. This is the first industry scale design, training and use of app vectorization.
UR - http://www.scopus.com/inward/record.url?scp=85006810397&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85006810397&partnerID=8YFLogxK
U2 - 10.1109/ASONAM.2016.7752297
DO - 10.1109/ASONAM.2016.7752297
M3 - Conference contribution
AN - SCOPUS:85006810397
T3 - Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016
SP - 599
EP - 606
BT - Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016
A2 - Kumar, Ravi
A2 - Caverlee, James
A2 - Tong, Hanghang
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 18 August 2016 through 21 August 2016
ER -