Predicting App Usage Based on Link Prediction in User-App Bipartite Network

Tan, Yaowen; Yu, Ke; Wu, Xiaofei; Pan, Di; Liu, Yang

doi:10.1007/978-3-319-73830-7_20

Cited by 5 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…TAN Yaowen et al constructed a User-App bipartite graph based on the network footprint data. This bipartite graph not only extracts the User-App correlation, but also expresses the similarity between users who used the same application [17]. Thus, an App usage prediction method based on link prediction was proposed.…”

Section: Related Work 21 Mobile App Start-up Predictionmentioning

confidence: 99%

Mobile App Start-Up Prediction Based on Federated Learning and Attributed Heterogeneous Network Embedding

et al. 2021

Future Internet

View full text Add to dashboard Cite

At present, most mobile App start-up prediction algorithms are only trained and predicted based on single-user data. They cannot integrate the data of all users to mine the correlation between users, and cannot alleviate the cold start problem of new users or newly installed Apps. There are some existing works related to mobile App start-up prediction using multi-user data, which require the integration of multi-party data. In this case, a typical solution is distributed learning of centralized computing. However, this solution can easily lead to the leakage of user privacy data. In this paper, we propose a mobile App start-up prediction method based on federated learning and attributed heterogeneous network embedding, which alleviates the cold start problem of new users or new Apps while guaranteeing users’ privacy.

show abstract

Section: Related Work 21 Mobile App Start-up Predictionmentioning

confidence: 99%

Mobile App Start-Up Prediction Based on Federated Learning and Attributed Heterogeneous Network Embedding

et al. 2021

Future Internet

View full text Add to dashboard Cite

show abstract

“…In addition, methods based on frequent sequence pattern mining are also used in some works for App usage prediction [17,18]. In [5], the authors construct the User-App Bipartite network based on the network footprint data and propose the App Usage prediction method based on link prediction. However, the prediction period was set to one day, which is far from practical, and the deep learning based on representation learning was also not involved.…”

Section: App Usage Predictionmentioning

confidence: 99%

“…Many previous studies have tackled the App prediction problem by means of network embedding, which can better mine the relationship between different types of applications and other information like time, location, etc. Tan et al [5] put forward extracting features and relationships from user-App bipartite. This bipartite not only describes the correlation between user and App but also expresses conjunctions of Apps used by a typical user or users who used the same application.…”

Section: Introductionmentioning

confidence: 99%

Graph-based Method for App Usage Prediction with Attributed Heterogeneous Network Embedding

Zhou

Liu

2020

Future Internet

View full text Add to dashboard Cite

Smartphones and applications have become widespread more and more. Thus, using the hardware and software of users’ mobile phones, we can get a large amount of personal data, in which a large part is about the user’s application usage patterns. By transforming and extracting these data, we can get user preferences, and provide personalized services and improve the experience for users. In a detailed way, studying application usage pattern benefits a variety of advantages such as precise bandwidth allocation, App launch acceleration, etc. However, the first thing to achieve the above advantages is to predict the next application accurately. In this paper, we propose AHNEAP, a novel network embedding based framework for predicting the next App to be used by characterizing the context information before one specific App being launched. AHNEAP transforms the historical App usage records in physical spaces to a large attributed heterogeneous network which contains three node types, three edges, and several attributes like App type, the day of the week. Then, the representation learning process is conducted. Finally, the App usage prediction problem was defined as a link prediction problem, realized by a simple neural network. Experiments on the LiveLab project dataset demonstrate the effectiveness of our framework which outperforms the three baseline methods for each tested user.

show abstract

“…They experiment on a number of real datasets and show that their proposed method is better than the previous methods. Tan et al (2018) focus on App usage prediction based on link prediction in bipartite networks. Their main task is to predict whether a user will use an App or not based on the historical NFP (Network footprint) data.…”

Section: Related Workmentioning

confidence: 99%

Real-Time Browsing Assistant on Web

Zuhori¹,

Miller²

2018

IJWI

View full text Add to dashboard Cite

Understanding user requirements based on their interactions with a website is becoming increasingly important. Hence, in this paper, a novel real-time navigation-support system is discussed. This system builds a personalized browsing assistant based on the current user request submitted to a web server. The process involves developing a behavior model using a Discrete Time Markov Chain (DTMCs) inference process. This is then used to monitor user activities, and thereafter suggest "where to go next". Finally, it updates the model in real time using a Markovian Decision Process (MDP). To evaluate the system, we provide a user study, case studies and conduct experiments on two datasets to verify the effectiveness of our proposed system.

show abstract

Predicting App Usage Based on Link Prediction in User-App Bipartite Network

Cited by 5 publications

References 14 publications

Mobile App Start-Up Prediction Based on Federated Learning and Attributed Heterogeneous Network Embedding

Mobile App Start-Up Prediction Based on Federated Learning and Attributed Heterogeneous Network Embedding

Graph-based Method for App Usage Prediction with Attributed Heterogeneous Network Embedding

Real-Time Browsing Assistant on Web

Contact Info

Product

Resources

About