Cross-Platform App Recommendation by Jointly Modeling Ratings and Texts

Cao, Da; He, Xiangnan; Nie, Liqiang; Wei, Xiaochi; Hu, Xia; Wu, Shunxiang; Chua, Tat-Seng

doi:10.1145/3017429

Cited by 88 publications

(43 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To evaluate the performance, we adopted root mean square error (RMSE), where a lower RMSE score indicates a be er performance. Note that RMSE has been widely used for evaluating regression tasks such as recommendation with explicit ratings [5,30] and click-through rate prediction [24]. We rounded up the prediction of each model to 1 or −1 if it was out of the range.…”

Section: Evaluation Protocolsmentioning

confidence: 99%

Neural Factorization Machines for Sparse Predictive Analytics

Chua

2017

Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval

Self Cite

1,126

788

View full text Add to dashboard Cite

Many predictive tasks of web applications need to model categorical variables, such as user IDs and demographics like genders and occupations. To apply standard machine learning techniques, these categorical predictors are always converted to a set of binary features via one-hot encoding, making the resultant feature vector highly sparse. To learn from such sparse data e ectively, it is crucial to account for the interactions between features.Factorization Machines (FMs) are a popular solution for e ciently using the second-order feature interactions. However, FM models feature interactions in a linear way, which can be insu cient for capturing the non-linear and complex inherent structure of real-world data. While deep neural networks have recently been applied to learn non-linear feature interactions in industry, such as the Wide&Deep by Google and DeepCross by Microso , the deep structure meanwhile makes them di cult to train.In this paper, we propose a novel model Neural Factorization Machine (NFM) for prediction under sparse se ings. NFM seamlessly combines the linearity of FM in modelling second-order feature interactions and the non-linearity of neural network in modelling higher-order feature interactions. Conceptually, NFM is more expressive than FM since FM can be seen as a special case of NFM without hidden layers. Empirical results on two regression tasks show that with one hidden layer only, NFM signi cantly outperforms FM with a 7.3% relative improvement. Compared to the recent deep learning methods Wide&Deep and DeepCross, our NFM uses a shallower structure but o ers be er performance, being much easier to train and tune in practice.

show abstract

Section: Evaluation Protocolsmentioning

confidence: 99%

Neural Factorization Machines for Sparse Predictive Analytics

Chua

2017

Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval

Self Cite

1,126

788

View full text Add to dashboard Cite

show abstract

“…As such, we can alleviate the time-consuming problem of ranking all items for each user during evaluation. In terms of evaluation metrics, we adopt Hit Ratio at rank k (HR@k) [16] and Normalized Discounted Cumulative Gain at rank k (NDCG@k) [1,5,12,14] to evaluate the performance of the ranked list generated by our models. In experimental parts we set k = 10 for both metrics.…”

Section: Experimental Settingsmentioning

confidence: 99%

Deep Item-based Collaborative Filtering for Top-N Recommendation

Xue

Wang

et al. 2019

ACM Trans. Inf. Syst.

Self Cite

267

133

View full text Add to dashboard Cite

Item-based Collaborative Filtering (short for ICF) has been widely adopted in recommender systems in industry, owing to its strength in user interest modeling and ease in online personalization. By constructing a user's profile with the items that the user has consumed, ICF recommends items that are similar to the user's profile. With the prevalence of machine learning in recent years, significant processes have been made for ICF by learning item similarity (or representation) from data. Nevertheless, we argue that most existing works have only considered linear and shallow relationship between items, which are insufficient to capture the complicated decision-making process of users.In this work, we propose a more expressive ICF solution by accounting for the nonlinear and higher-order relationship among items. Going beyond modeling only the second-order interaction (e.g., similarity) between two items, we additionally consider the interaction among all interacted item pairs by using nonlinear neural networks. Through this way, we can effectively model the higher-order relationship among items, capturing more complicated effects in user decision-making. For example, it can differentiate which historical itemsets in a user's profile are more important in affecting the user to make a purchase decision on an item. We treat this solution as a deep variant of ICF, thus term it as DeepICF. To justify our proposal, we perform empirical studies on two public datasets from MovieLens and Pinterest. Extensive experiments verify the highly positive effect of higher-order item interaction modeling with nonlinear neural networks. Moreover, we demonstrate that by more fine-grained second-order interaction modeling with attention network, the performance of our DeepICF method can be further improved.

show abstract

“…The study [36] was generally focused on recommending independent items to users who were suggested by a hybrid cross-platform app recommendation (STAR) system. Another study [37] introduced recommender systems on mobile platforms based on user profiles generated from the installed apps.…”

Section: Application Recommender System Studiesmentioning

confidence: 99%

A Recommender System for Mobile Applications of Google Play Store

Fuad¹,

Bayoumi²,

Al-Yahya³

2020

IJACSA

View full text Add to dashboard Cite

With the growth in the smartphone market, many applications can be downloaded by users. Users struggle with the availability of a massive number of mobile applications in the market while finding a suitable application to meet their needs. Indeed, there is a critical demand for personalized application recommendations. To address this problem, we propose a model that seamlessly combines content-based filtering with application profiles. We analyzed the applications available on the Google Play app store to extract the essential features for choosing an app and then used these features to build app profiles. Based on the number of installations, the number of reviews, app size, and category, we developed a content-based recommender system that can suggest some apps for users based on what they have searched for in the application's profile. We tested our model using a k-nearest neighbor algorithm and demonstrated that our system achieved good and reasonable results.

show abstract

Cross-Platform App Recommendation by Jointly Modeling Ratings and Texts

Cited by 88 publications

References 58 publications

Neural Factorization Machines for Sparse Predictive Analytics

Neural Factorization Machines for Sparse Predictive Analytics

Deep Item-based Collaborative Filtering for Top-N Recommendation

A Recommender System for Mobile Applications of Google Play Store

Contact Info

Product

Resources

About