Search

The user perception of mobile game is crucial for improving user experience and thus enhancing game profitability. The sparse data captured in the game can lead to sporadic performance of the model. This paper proposes a new method, the balanced graph factorization machine (BGFM), based on existing algorithms, considering the data imbalance and important high-dimensional features. The data categories are first balanced by Borderline-SMOTE oversampling, and then features are represented naturally in a graph-structured way. The highlight is that the BGFM contains interaction mechanisms for aggregating beneficial features. The results are represented as edges in the graph. Next, BGFM combines factorization machine (FM) and graph neural network strategies to concatenate any sequential feature interactions of features in the graph with an attention mechanism that assigns inter-feature weights. Experiments were conducted on the collected game perception dataset. The performance of proposed BGFM was compared with eight state-of-the-art models, significantly surpassing all of them by AUC, precision, recall, and F-measure indices. [ABSTRACT FROM AUTHOR]

The subject of personalized exercise recommendation holds significant relevance within the domain of personalized services in smart education. Nevertheless, traditional algorithms have often lacked a deep understanding of student characteristics and failed to adequately explore the relationship between knowledge mastery and question-answering behaviors, leading to low recommendation accuracy. To address these issues, combining the knowledge tracing machine and the user-based collaborative filtering algorithm, as a KTM-based multi-feature fusion exercise recommendation model, SKT-MFER was proposed. Firstly, as a knowledge tracking model, KTM-LC, incorporating student learning behaviors and learning abilities, was constructed to accurately assess the student’s knowledge mastery level. Subsequently, two filters were implemented to ensure the exercise recommendation’s accuracy: the first was an initial screening utilizing the knowledge point mastery matrix to eliminate students who were similar to the target student, and the second was a filtering process considering the combined similarity of cognitive state similarity and exercise difficulty similarity. Through extensive experiments, it proves that the proposed method yields better results than some existing baseline models.

The neural network-enhanced factorization machine (FM) model, which can capture more high-order feature interactions and improve the accuracy of predictions, has become a research hotspot in the field of recommendation algorithms. Aiming at the problem that existing models do not comprehensively consider high-order interaction features and original low-order features when modeling the interactions between users and items, and in order to improve the model's ability to model user preferences, this paper proposes a new deep neural network-enhanced pairwise bilinear factorization machine model, DeepPRBFM, by combining depth neural networks and pairwise learning. This model adopts a bilinear structure with a pair of inputs containing positive and negative samples, utilizes multi-layer ResNet to preserve low-order features, enhances the interaction of high-order features with DNN, and employs a pairwise ranking-based loss function. Moreover, in the bilinear structure, increasing the proportion of negative samples can not only significantly alleviate the cold start problem of the recommendation system but also improve the prediction performance of the model. Experiments conducted on two realworld datasets show that the proposed model achieves higher recommendation accuracy and outperforms other models in objective metrics such as HR and NDCG. [ABSTRACT FROM AUTHOR]

The user perception of mobile game is crucial for improving user experience and thus enhancing game profitability. The sparse data captured in the game can lead to sporadic performance of the model. This paper proposes a new method, the balanced graph factorization machine (BGFM), based on existing algorithms, considering the data imbalance and important high-dimensional features. The data categories are first balanced by Borderline-SMOTE oversampling, and then features are represented naturally in a graph-structured way. The highlight is that the BGFM contains interaction mechanisms for aggregating beneficial features. The results are represented as edges in the graph. Next, BGFM combines factorization machine (FM) and graph neural network strategies to concatenate any sequential feature interactions of features in the graph with an attention mechanism that assigns inter-feature weights. Experiments were conducted on the collected game perception dataset. The performance of proposed BGFM was compared with eight state-of-the-art models, significantly surpassing all of them by AUC, precision, recall, and F-measure indices.

The integration of graph embedding technology and collaborative filtering algorithms has shown promise in enhancing the performance of recommendation systems. However, existing integrated recommendation algorithms often suffer from feature bias and lack effectiveness in personalized user recommendation. For instance, users' historical interactions with a certain class of items may inaccurately lead to recommendations of all items within that class, resulting in feature bias. Moreover, accommodating changes in user interests over time poses a significant challenge. This study introduces a novel recommendation model, RCKFM, which addresses these shortcomings by leveraging the CoFM model, TransR graph embedding model, backdoor tuning of causal inference, KL divergence, and the factorization machine model. RCKFM focuses on improving graph embedding technology, adjusting feature bias in embedding models, and achieving personalized recommendations. Specifically, it employs the TransR graph embedding model to handle various relationship types effectively, mitigates feature bias using causal inference techniques, and predicts changes in user interests through KL divergence, thereby enhancing the accuracy of personalized recommendations. Experimental evaluations conducted on publicly available datasets, including "MovieLens-1M" and "Douban dataset" from Kaggle, demonstrate the superior performance of the RCKFM model. The results indicate a significant improvement of between 3.17% and 6.81% in key indicators such as precision, recall, normalized discount cumulative gain, and hit rate in the top-10 recommendation tasks. These findings underscore the efficacy and potential impact of the proposed RCKFM model in advancing recommendation systems. [ABSTRACT FROM AUTHOR]

The proliferation of internet communication channels has increased telecom fraud, causing billions of euros in losses for customers and the industry each year. Fraudsters constantly find new ways to engage in illegal activity on the network. To reduce these losses, a new fraud detection approach is required. Telecom fraud detection involves identifying a small number of fraudulent calls from a vast amount of call traffic. Developing an effective strategy to combat fraud has become challenging. Although much effort has been made to detect fraud, most existing methods are designed for batch processing, not real-time detection. To solve this problem, we propose an online fraud detection model using a Neural Factorization Autoencoder (NFA), which analyzes customer calling patterns to detect fraudulent calls. The model employs Neural Factorization Machines (NFM) and an Autoencoder (AE) to model calling patterns and a memory module to adapt to changing customer behaviour. We evaluate our approach on a large dataset of real-world call detail records and compare it with several state-of-the-art methods. Our results show that our approach outperforms the baselines, with an AUC of 91.06%, a TPR of 91.89%, an FPR of 14.76%, and an F1-score of 95.45%. These results demonstrate the effectiveness of our approach in detecting fraud in real-time and suggest that it can be a valuable tool for preventing fraud in telecommunications networks.

The growth of music streaming market is expected to be boosted by the rising use of smart devices and the streaming platforms in the forecast period. Music recommendation systems for groups have been intensively studied as the application scenario becomes more dynamic, where different kinds of group formulations lead to different levels of similarity in music tastes. This paper proposes a music recommendation framework to generate a Top-N list in high-similar, high-dissimilar, and regular groups. Additionally, we investigate the performance of the recommendation system in a wide range of aspects, including precision, ranking quality, and novelty. Based on the listening frequency of each user and for each track, implicit feedback is also derived from the track popularity. To prevent popularity bias, we combine track popularity and group popularity into latent factor models to improve the performance of the recommendation algorithm. As a result of our evaluation, we found that the proposed methods provided excellent recommendations regarding accuracy, diversity, and novelty. [ABSTRACT FROM AUTHOR]

SetRank is a novel Setwise Bayesian collaborative ranking model that can model implicit feedback data more closely to real-world scenarios. However, SetRank only considers collaborative information and lacks effective use of content information. In order to solve the above problems, a factorization machine model based on Setwise ranking, named SRFMs, is proposed. Drawing on the Setwise ranking proposed in SetRank to solve the problem of missing negative samples in implicit feedback, a factorization machine is chosen as a predictor to model content information, and model user preferences from the perspective of optimal item ranking. Furthermore, to improve the fixed feature representation and the lack of higher-order feature interactions in standard FM, inspired by IFM, SR-DIFM model is constructed to incorporate the SRFMs by combining input-aware networks with neural networks. Experimental results on two real-world datasets demonstrate that the proposed model outperforms the state-of-art model in terms of evaluation metrics including HR, NDCG and mAP, and can improve the accuracy of recommendations which make better use of user and item content information while solving the recommendation problem under implicit feedback. [ABSTRACT FROM AUTHOR]

In the context of the rapid advancement of the Industrial Internet and Urban Internet, a crucial trend is emerging in the realization of unified, service-oriented, and componentized encapsulation of IT and OT heterogeneous entities underpinned by a service-oriented architecture. This is pivotal for achieving componentized construction and development of extensive industrial software systems. In addressing the diverse demands of application tasks, the efficient and precise recommendation of service components has emerged as a pivotal concern. Existing recommendation models either focus solely on low-order interactions or emphasize high-order interactions, disregarding the distinction between implicit and explicit aspects within high-order interactions as well as the integration of high-order and low-order interactions. This oversight leads to subpar accuracy in recommendations. Real-world data exhibit intricate structures and nonlinearity. In practical applications, different interaction components exhibit varying predictive capabilities. Therefore, in this paper we propose an EIAFM model that fuses explicit and implicit higher-order feature interactions and introduce an attention mechanism to identify which low-level feature interactions contribute more significantly to the prediction results. This approach leads to increased interpretability, combining both generalization and memory capabilities. Through comprehensive experiments on authentic datasets that align with the characteristics of the Service Recommendation of Industrial Software Components problem, we demonstrate that the EIAFM model excels compared to other cutting-edge models in terms of recommendation effectiveness, with the evaluation metrics for the AUC and log-loss reaching values of 0.9281 and 0.3476, respectively. [ABSTRACT FROM AUTHOR]

To date, various optimization algorithms have been used to design non-intuitive photonic structures with unconventional optical performance. Good training datasets facilitate the optimization process, particularly when an objective function has a non-convex shape containing multiple local optima in a continuous parametric space. Herein, we developed a discrete-to-continuous optimization algorithm and confirmed its validity by designing and fabricating deep-ultraviolet antireflective MgF2/LaF3 multilayers. For discrete optimization, a multilayer was encoded into a binary vector with multiple bits; a 10 nm thick MgF2 or LaF3 layer was assigned a binary digit of 0 or 1, respectively. Using the binary-based training datasets, a factorization machine formulated a surrogate function, which discovered the ground binary vector representing a near-optimal figure of merit. Then, the figure of merit was refined through continuous optimization (e.g., using an interior-point method) of the ground binary vector. MgF2/LaF3 multilayers with a variety of bit levels were created to attain a minimum average angular (0°–45°) reflectance at 193 nm. A MgF2/LaF3 multilayer optimized at ten bits (i.e., a total thickness of approximately 100 nm) yielded an average reflectance of 0.2%, which agreed well with the experimental results. Moreover, an integrated ray-wave optics simulation predicted that a single CaF2 plano-convex lens coated with the optimized multilayer could exhibit a transmittance of 99.7%. The developed optimization approach will be widely applicable to any photonic structures that can represent a binary vector with multiple bits, such as microwave metasurfaces, in addition to being useful for producing ideal optical multilayers.

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Due to its automatic feature learning ability and high performance, deep learning has gradually become the mainstream of artificial intelligence in recent years, playing a role in many fields. Especially in the medical field, the accuracy rate of deep learning even exceeds that of doctors. This paper introduces several deep learning algorithms: Artificial Neural Network (NN), FM-Deep Learning, Convolutional NN and Recurrent NN, and expounds their theory, development history and applications in disease prediction; we analyze the defects in the current disease prediction field and give some current solutions; our paper expounds the two major trends in the future disease prediction and medical field—integrating Digital Twins and promoting precision medicine. This study can better inspire relevant researchers, so that they can use this article to understand related disease prediction algorithms and then make better related research. [ABSTRACT FROM AUTHOR]

Recommender system can find information that satisfies the user′s individual needs from huge amount of information. With the development of deep learning, deep learning has been widely applied in recommender systems. CTR prediction plays an important role in recommender system and has been widely used in many fields such as personalized recommendation, information retrieval, online advertising and so on. For the issue of large and sparse data in recommender system, this study merges xDeepFM model with attention network, and proposes a new CTR prediction model based on deep learning, which is called Atte-xDeepFM model. This model can solve the issue of feature scarcity, effectively learn the interactions relationship between features, and does not need to manually extract useful information in feature engineering. The comparative experiments on Avazu and Criteo data sets prove the effectiveness of the proposed model. Compared with the algorithm model commonly used in CTR prediction, the proposed model has better recommendation effect. [ABSTRACT FROM AUTHOR]

Game user perception is of great significance for game developers and network operators for improving service quality and operational efficiency. At present, the most common approach is to use the linear model that considers only the impact of network factors evaluation on user perception. The interpretation process is complex and useful, but invisible feature interaction data are not taken into account. As a result, user perception evaluation can only be interpreted by experienced experts, which is both time-consuming and laborious. In this paper, aiming at the shortcomings of existing algorithms, a location–time-aware factorization machine model (LTFM) is proposed by exploiting the location projection and time projection of users and services and fuzzy set theory. Our proposed LTFM can be decomposed into two parts: first, an original game quality of experience (QoE) dataset is extended. LTFM uses location and time information to map to latent vectors, which increases the number of records in each game data, involving no additional information. Then, LTFM utilizes fuzzy set theory to strengthen the positive feature interactions and reduce the negative feature interactions. The factorization machine is used to mine a number of potential features in the user's invoking service behavior. The multiplayer online battle arena (MOBA) game perception dataset is obtained with reference to the ITU-T standard to verify the advanced nature of the proposed model. Experimental results show that LTFM outperforms existing algorithms in terms of prediction accuracy and model interpretability. Not only can accurate user experience quality categories be produced, but also the impact of individual characteristics and their feature interactions can be explained, which helps operators to make better optimization decisions. [ABSTRACT FROM AUTHOR]

In the problem of unified classroom performance prediction, there is a certain lag in the prediction, and there are also problems such as the data sparsity and single feature in the data. In addition, feature engineering is often carried out manually in modeling, which highly depends on the professional knowledge and experience of engineers and affects the accuracy of the prediction to a certain extent. To solve the abovementioned gaps, we proposed an online course score prediction model with a high time efficiency that combines multiple features. The model uses a deep neural network, which can automatically carry out feature engineering and reduce the intervention of artificial feature engineering, thus significantly improving the time efficiency. Secondly, the model uses a factorization machine and two kinds of neural networks to consider the influence of first-order features, second-order features, and higher-order features at the same time, and it fully learns the relationship between the features and scores, which improves the prediction effect of the model compared to using only single feature learning. The performance of the model is evaluated on the learning analysis dataset from Fall 2015 to Spring 2021 and includes 412 courses with 600 students. The experimental results show that the performance of the prediction model based on the feature combination proposed in the present study is better than the previous performance prediction model. More importantly, our model has the best time efficiency of below 0.3 compared to the other models. [ABSTRACT FROM AUTHOR]

Machine learning algorithms are susceptible to cyberattacks, posing security problems in computer vision, speech recognition, and recommendation systems. So far, researchers have made great strides in adopting adversarial training as a defensive strategy. Single-step adversarial training methods have been proposed as viable solutions for improving model generality and resilience. However, there has been little study to address this issue in the context of ownership-based recommendations, which may fail to capture stable results. In this work, we adapt the single-step adversarial training for ownership recommendation systems. Our main technical contributions are as follows: (1) We propose Adversarial Consumption and Production Relationship (ACPR), a model that combines factorization machine and single-step adversarial training for ownership recommendations. It enables us to take advantage of modeling consumption-production interactions with a factorization machine instead of the conventional matrix factorization method for ownership recommendations. (2) We enrich the ACPR technique with directional adversarial training and call our technique Adversarial Consumption and Production Relationship-Aware Directional Adversarial Model (ACPR-ADAM). The idea behind our ACPR-ADAM is that instead of the worst perturbation direction, the perturbation direction in the embedding space is restricted to other examples in the current embedding space, allowing us to incorporate the collaborative signal into the training process. Lastly, through extensive evaluations on Reddit and Pinterest, we demonstrate that our proposed method outperforms state-of-the-art methods. Compared with CPR and ACPR on Reddit and Pinterest datasets, our proposed ACPR-ADAM achieves 93%, 88%, and 72%, 69% enhancement in terms of AUC and HR, respectively.

The rating data of recommendation systems are too sparse, and many existing studies introduce review information to alleviate this phenomenon, but the connection between reviews and the importance of each review is not fully considered in the feature extraction of the information of reviews. To address this problem, this paper proposes the Local-Global Awareness Attention Model (LGAA) to model the comment information. Through the local attention mechanism, the neighborhood attention score is calculated by fully considering the association between each comment and other comments; then the global attention mechanism is used to calculate the global attention score of each comment and weight the sum of all comments, and fuse the scoring information to form a feature vector; finally, it is fed into a factorization machine and a fully connected network for scoring prediction. The results show that LGAA can achieve better performance than the benchmark algorithm.

In the context of the rapid advancement of the Industrial Internet and Urban Internet, a crucial trend is emerging in the realization of unified, service-oriented, and componentized encapsulation of IT and OT heterogeneous entities underpinned by a service-oriented architecture. This is pivotal for achieving componentized construction and development of extensive industrial software systems. In addressing the diverse demands of application tasks, the efficient and precise recommendation of service components has emerged as a pivotal concern. Existing recommendation models either focus solely on low-order interactions or emphasize high-order interactions, disregarding the distinction between implicit and explicit aspects within high-order interactions as well as the integration of high-order and low-order interactions. This oversight leads to subpar accuracy in recommendations. Real-world data exhibit intricate structures and nonlinearity. In practical applications, different interaction components exhibit varying predictive capabilities. Therefore, in this paper we propose an EIAFM model that fuses explicit and implicit higher-order feature interactions and introduce an attention mechanism to identify which low-level feature interactions contribute more significantly to the prediction results. This approach leads to increased interpretability, combining both generalization and memory capabilities. Through comprehensive experiments on authentic datasets that align with the characteristics of the Service Recommendation of Industrial Software Components problem, we demonstrate that the EIAFM model excels compared to other cutting-edge models in terms of recommendation effectiveness, with the evaluation metrics for the AUC and log-loss reaching values of 0.9281 and 0.3476, respectively.

Under variable working conditions, the tool status signal is affected by changing machine processing parameters, resulting in a decreased prediction accuracy of the remaining useful life (RUL). Aiming at this problem, a method based on multi-sensor fusion for tool RUL prediction was proposed. Firstly, the factorization machine (FM) was used to extract the nonlinear processing features in the low-frequency condition signal, and the one-dimensional separable convolution was applied to extract tool life state features from multi-channel high-frequency sensor signals. Secondly, the residual attention mechanism was introduced to weight the low-frequency condition characteristics and high-frequency state characteristics, respectively. Finally, the features extracted in the low-frequency and high-frequency parts were input into the full connection layer to integrate working condition information and state information to suppress the influence of variable conditions and improve prediction accuracy. The experimental results demonstrated that the method could predict the remaining life of the tool effectively, and the accuracy and stability of the model are better than several other methods. [ABSTRACT FROM AUTHOR]

Controller area network (CAN) is the most commonly used bus technology for In-vehicle network and uses multicast communication without corresponding security measures. Therefore, the message data field is vulnerable to tampering and other attacks. Recent machine learning-based intrusion detection methods for CAN bus messages only use the information contained in the message data field and do not take into account the contribution made by the neighboring information of CAN bus messages. In addition, previous models considered the data domain information of CAN bus messages as separate features and did not consider the unique weight of each feature, as well as the second-order interaction information between features. Therefore, we propose a novel intrusion detection model, The Hybrid Similar Neighborhood Robust Factorization Machine Model (HSNRFM), for detecting anomalies in CAN bus messages to address the shortcomings and problems of the previous models. To be able to incorporate the contribution of similar neighborhood information and learn the unique weight parameters of each feature in the model decision process, as well as additional second-order interaction information between features, the HSNRFM model solves the above problem using a similarity calculation method and a factorization machine model. Comprehensive experimental results are compared on real vehicle datasets. The HSNRFM model has AUC values of 0.9216 and 0.901 and AUPR values of 0.9194 and 0.9018 on two real datasets, respectively. And the results show that our proposed HSNRFM model has excellent detection efficiency for intrusion detection of CAN bus messages. [ABSTRACT FROM AUTHOR]

In order to apply the recommendation system to the field of military training and give full play to the value of military training big data in personalized training, a hybrid recommendation model called Cross-DeepFM is proposed. Firstly, the real military training data are collected and preprocessed, and the custom military training dataset is constructed. Then, the structure of the Cross-DeepFM model is designed by combining the deep residual neural network, deep cross network and factorization machine, and the details of the model are analyzed. Finally, the comparison and analysis are carried out on the custom military training data set. The experimental results show that the proposed model is more accurate than the mainstream recommendation model and can effectively complete the personalized recommendation task of military training. [ABSTRACT FROM AUTHOR]

Factorization Machines (FMs) are effective in incorporating side information to overcome the cold-start and data sparsity problems in recommender systems. Traditional FMs adopt the inner product to model the second-order interactions between different attributes, which are represented via feature vectors. The problem is that the inner product violates the triangle inequality property of feature vectors. As a result, it cannot well capture fine-grained attribute interactions, resulting in sub-optimal performance. Recently, the euclidean distance is exploited in FMs to replace the inner product and has delivered better performance. However, previous FM methods including the ones equipped with the euclidean distance all focus on the attribute-level interaction modeling, ignoring the critical intrinsic feature correlations inside attributes. Thereby, they fail to model the complex and rich interactions exhibited in the real-world data. To tackle this problem, in this paper, we propose a FM framework equipped with generalized metric learning techniques to better capture these feature correlations. In particular, based on this framework, we present a Mahalanobis distance and a deep neural network (DNN) methods, which can effectively model the linear and non-linear correlations between features, respectively. Besides, we design an efficient approach for simplifying the model functions. Experiments on several benchmark datasets demonstrate that our proposed framework outperforms several state-of-the-art baselines by a large margin. Moreover, we collect a new large-scale dataset on second-hand trading to justify the effectiveness of our method over cold-start and data sparsity problems in recommender systems. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Chongqing University of Posts & Telecommunications (Natural Science Edition) is the property of Chongqing University of Posts & Telecommunications and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Aiming at the problems of single feature combination of recommendation model, resolution of a large amount of valuable feature information, and over-fitting in deep learning, a new attentional scoring mechanism called attention capsule was designed, and a deep factorization machine model based on attention capsule was proposed.Users’ historical clicking and candidate items were processed through weight calculation based on the DeepFM model, reducing the impact of irrelevant features on the model, and the differential impact of different historical behaviors on users’ interests was fully explored.The adaptive regularization formulation was added to the training, which effectively reduced over-fitting without affecting the training speed.The comparison test on two public data sets shows that the proposed model is significantly enhanced in loss function and GAUC compared to other models.

The era of everythingasaservice led to an explosion of services with similar functionalities on the internet. Quickly obtaining a high-quality service has become a research focus in the field of service recommendation. Studies show that quality of service (QoS) predictions are an effective way to discover services with high quality. However, sparse data and performance fluctuation challenge the accuracy and robustness of QoS prediction. To solve these two challenges, this paper proposes a location-based deep factorization machine model, namely LDFM, by employing information entropy and location projection of users and services. Particularly, our LDFM can be decomposed into three phases: i) extending a raw QoS dataset without introducing additional information, where LDFM projects the existing users (services) in the direction of their position vectors to increase the number of users (services) as well as the number of records that users invoke services; ii) mining a sufficient number of potential features behind the behaviors of users who invoke services, where LDFM employs a factorization machine to extract potential features of breadth with low dimensions (i.e., one and two dimensions) and utilizes deep learning to seek potential depth features with high dimensions; and iii) weighting extracted features within various dimensions, where LDFM employs information entropy to strengthen the positive effects of valid features while reducing the negative impacts generated by biased features. Our experimental results (including t-test analyses) show that our proposed LDFM always performs well under different user-service matrix densities and performs better than existing start-of-the-art methods in terms of the accuracy and robustness of QoS predictions. [ABSTRACT FROM AUTHOR]

In recent years, the recommendation system and robot learning are undoubtedly the two most popular application fields, and the core algorithms supporting these two fields are deep learning based on perception and reinforcement learning based on exploration learning, respectively. How to combine these two fields to better improve the development of the whole machine learning field is the dream of numerous researchers. The Deep Reinforcement Network (DRN) model successfully embedded reinforcement learning into the recommendation system, which provided a good idea for subsequent researchers. However, the disadvantage is also obvious, that is, the DRN model is built for news recommendations, meaning that the DRN model is not transferable, which is also the defect of many current recommendation system models. Meanwhile, the agent learning method adopted by the DRN model is primitive and inefficient. Among many models and algorithms that have emerged in recent years, we use the newly proposed deployment efficiency to measure their comprehensive quality and found that few models focus on both efficiency and performance improvement. To fill the gap of model deployment efficiency neglected by many researchers and to create a model of reinforcement learning agents with stronger performance, we have been exploring and trying to complete research on the Gate Attentional Factorization Machines (GAFM) model. Finally, we successfully integrated the GAFM model and reinforcement learning. The Deep Reinforcement Factorization Machines (DRFM) model proposed in this paper is based on the combination of deep learning with high perception ability and reinforcement learning with high exploration ability, centered on improving the deployment efficiency and learning performance of the model. The GAFM model is modified and upgraded using multidisciplinary techniques, and a new model-based random exploration strategy is proposed to update and optimize the recommendation list efficiently. Through parallel contrast experiments on various datasets, it is proved that the DRFM model surpasses the traditional recommendation system model in all aspects. The DRFM model is far superior to other models in terms of performance and robustness, and also significantly improved in terms of deployment efficiency. At the same time, we conduct a comparative analysis with the latest deep reinforcement learning algorithm and prove the unique advantages of the DRFM model. [ABSTRACT FROM AUTHOR]

Recently, recommender systems are applied to provide personalized recomendation for healthcare wearables. However, due to the sparsity problem, traditional recommendation algorithms are difficult to achieve desired performance. Considering that consumers often buy and rate other types of items on E-commerce platforms, we can leverage significant information in the auxiliary domains to improve the recommendation performance of healthcare wearables, which can be regarded as cross-domain recommendation. However, traditional cross-domain recommendation model cannot fully represent user’s characteristics and fail to consider the leaks of original auxiliary domain ratings during the information transfer process. To overcome the two shortcomings, this paper proposes a Privacy-Preserving Cross-Domain Healthcare Wearables Recommendation algorithm (PPCDHWRec). Firstly, user’s characteristics are divided into domain-dependent features and domain-independent features, which complement each other and fully depict the user’s characteristics. Secondly, inspired by the latent factor model, we factorize the original rating information of each auxiliary domain by Funk-SVD and Orthogonal Nonnegative Matrix Tri-Factorization (ONMTF) model, to obtain user’s domain-dependent and domain-independent features, respectively. Finally, the Factorization Machine algorithm is used to fuse the obtained user’s features with the target domain information to provide the recommendation results. By hiding the item latent factors obtained in the factorization process, PPCDHWRec ensures that the original information cannot be inferred from the transferred user hidden vector. Hence, PPCDHWRec is a privacy-preserving recommendation model. Experiments on two groups of auxiliary domains, having high and low correlations with target domain, show the effectiveness of PPCDHWRec. [ABSTRACT FROM AUTHOR]

In recent years, with the increase of computer computing power, Deep Learning has begun to be favored. Its learning of non-linear feature combinations has played a role that traditional machine learning cannot reach in almost every field. The application of Deep Learning has also driven the advancement of Factorization Machine (FM) in the field of recommendation systems, because Deep Learning and FM can learn high-order and low-order features combinations respectively, and FM's hidden vector system enables it to learn information from sparse data. The integration of them has attracted the attention of many scholars. They have researched many classic models such as Factorization-supported Neural Network (FNN), Product-based Neural Networks (PNN), Inner PNN (IPNN), Wide&Deep, Deep&Cross, DeepFM, etc. for the Click-Through-Rate (CTR) problem, and their performance is getting better and better. This kind of model is also suitable for agriculture, meteorology, disease prediction and other fields due to the above advantages. Based on the DeepFM model, we predicts the incidence of hepatitis in each sample in the structured disease prediction data of the 2020 Artificial Intelligence Challenge Preliminary Competition, and make minor improvements and parameter adjustments to DeepFM. Compared with other models, the improved DeepFM has excellent performance in AUC. This research can be applied to electronic medical records to reduce the workload of doctors and make doctors focus on the samples with higher predicted incidence rates. For some changing data, such as blood pressure, height, weight, cholesterol, etc., we can introduce the Internet of Medical Things (IoMT). IoMT's sensors can be used to conduct transmission to ensure that the disease can be predicted in time, just in case. After joining IoMT, a healthcare system is formed, which is superior in forecasting and time performance.

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Factorization machine (FM) has been widely applied in recommender systems since its publication, and to capture feature interactions, FM represents its 2nd-order coefficient of any two features as inner product of the corresponding embedding vectors in Euclidean space. Considering that objects in recommender systems such as items, users, properties and contexts can be described as a heterogeneous network exhibiting hierarchical structures, whereas flat Euclidean space is not able to capture this kind of structure, restricting the feature representation ability of FM, this paper proposes hyperbolic FM (HFM). It represents each feature as a vector in hyperbolic space rather than in Euclidean space, and evaluates the 2nd-order feature interaction strength with hyperbolic distance measure. The reason for adopting hyperbolic geometry is that it has been shown to be the underlying embedding space of hierarchical structures, like trees, graphs and vocabulary. This paper designs two HFMs based on Poincaré ball model and hyperboloid model, respectively, and derives the corresponding Riemannian gradient descent algorithm for optimization. Experiments conducted on various datasets indicate that HFM achieves better performance than original FM with identical number of trainable parameters, and reveals the hierarchical structure of features which is missing in FM, offering explanability to some extent.

Knowledge tracing (KT) aims to closely trace the knowledge level of students during their learning. KT is often implemented by intelligent tutoring systems (ITS) to predict the student performance, and schedule the individual learning plan for each student. However, most existing KT models cannot predict the knowledge forgetting well due to the lack of temporal information and some KT models for knowledge forgetting prediction are not accurate enough. To resolve these issues, this paper proposes a recurrent knowledge tracing machine (RKTM), which temporally enriches the encoding of knowledge tracing machine (KTM) and difficulty, student ability, skill, and student skill practice history (DAS3H) with the knowledge state of students. RKTM consists of two major components, including the tracing component and the predicting component. The tracing component temporally traces the knowledge state of a student, while the predicting component captures the interaction between the current learning scenario and the current knowledge state of that student and provides accurate prediction of student performance. Experiments show that the proposed RKTM well outperforms KTM, DAS3H, and some other models. [ABSTRACT FROM AUTHOR]

Prediction of potential microRNA-disease associations is one of the important tasks in computational biology fields. Mining more sophisticated features can improve the performance of the prediction methods. This article proposes a novel algorithm (ISFMDA) that can effectively learn low- or high-order interactions of recursive feature elimination selected features by an extreme gradient boosting, a factorization machine, and a deep neural network. As a result, ISFMDA can obtain an area under receiver operating characteristic curve (AUROC) of 0.9342 ± 0.0007 in fivefold cross-validation tests with 51.25% of original features, which verifies the effectiveness of the methods. [ABSTRACT FROM AUTHOR]

The rating data of recommendation systems are too sparse, and many existing studies introduce review information to alleviate this phenomenon, but the connection between reviews and the importance of each review is not fully considered in the feature extraction of the information of reviews. To address this problem, this paper proposes the Local-Global Awareness Attention Model (LGAA) to model the comment information. Through the local attention mechanism, the neighborhood attention score is calculated by fully considering the association between each comment and other comments; then the global attention mechanism is used to calculate the global attention score of each comment and weight the sum of all comments, and fuse the scoring information to form a feature vector; finally, it is fed into a factorization machine and a fully connected network for scoring prediction. The results show that LGAA can achieve better performance than the benchmark algorithm. [ABSTRACT FROM AUTHOR]

The wide adoption of Service Oriented Architecture (SOA) has driven the creation of a massive amount of applications on the Internet, which includes the popular Mashups composed from multiple existing Web APIs. The availability of a large number of Web APIs with diverse functionalities on the Web makes it difficult for users to find APIs meeting their needs for Mashup development. To relieve this difficulty, recommending Web APIs for Mashup development has become an effective solution. A dozen of service recommendation approaches were proposed based on multi-dimensional features extracted from the service repository over the last couple of years, e.g., similarity based matching methods, matrix factorization based models, and factorization machine based models. Among these existing works, Factorization Machine (FM) based models, in particular the deep learning based FM models, have shown better performance compared with other conventional collaborative filtering techniques. Despite their superiority, the deep learning based FMs still have some strong model assumptions that can harm the recommendation accuracy. For example, it models factorized interactions with the same weight and ignores the non-linear and complex inherent structure in data. In a real-world service recommendation scenario, different predictor variables usually have different predictive power and not all features are predictable for estimating the target. Also, higher-order feature interactions are usually underlain in complex user-service environments. To address these deficiencies, this paper proposes a hybrid factorization machine model with a novel neural network architecture, named NAFM, which integrates a deep neural network to capture the non-linear and complex feature interactions and uses an attention mechanism to capture the varying importance of feature interactions. Comprehensive experiments are conducted on a real-world dataset from ProgrammableWeb. The experimental results show that the proposed approach outperforms the existing state-of-the-art models for service recommendation. [ABSTRACT FROM AUTHOR]

Game user perception is of great significance for game developers and network operators for improving service quality and operational efficiency. At present, the most common approach is to use the linear model that considers only the impact of network factors evaluation on user perception. The interpretation process is complex and useful, but invisible feature interaction data are not taken into account. As a result, user perception evaluation can only be interpreted by experienced experts, which is both time-consuming and laborious. In this paper, aiming at the shortcomings of existing algorithms, a location–time-aware factorization machine model (LTFM) is proposed by exploiting the location projection and time projection of users and services and fuzzy set theory. Our proposed LTFM can be decomposed into two parts: first, an original game quality of experience (QoE) dataset is extended. LTFM uses location and time information to map to latent vectors, which increases the number of records in each game data, involving no additional information. Then, LTFM utilizes fuzzy set theory to strengthen the positive feature interactions and reduce the negative feature interactions. The factorization machine is used to mine a number of potential features in the user’s invoking service behavior. The multiplayer online battle arena (MOBA) game perception dataset is obtained with reference to the ITU-T standard to verify the advanced nature of the proposed model. Experimental results show that LTFM outperforms existing algorithms in terms of prediction accuracy and model interpretability. Not only can accurate user experience quality categories be produced, but also the impact of individual characteristics and their feature interactions can be explained, which helps operators to make better optimization decisions.