Search

As the second largest bond market in the world, China's bond market has attracted extensive attention in recent years. Given its importance in facilitating financing arrangements and informing investment decisions, accurate bond coupon prediction is valuable. This paper proposes an ensemble model combining TabNet, DeepFM, and XGBoost for predicting the coupons of investment-grade corporate bonds. Specifically, to optimize the hyperparameters of the proposed model, an improved butterfly optimization algorithm incorporating the concepts of good point sets, refraction opposition-based learning, switching probability adjustment, and Solis & Wets search strategies is developed. Extensive experiments using data on China's investment-grade corporate bonds demonstrate the superior performance of the proposed model in the accuracy of bond coupon predictions. Additionally, the importance of various features has been discussed. The results show that the base interest rate for valuation and term to maturity are important to bond coupon predictions obtained by the proposed model. [ABSTRACT FROM AUTHOR]

Assessing the efficacy of algorithms plays a pivotal role in advancing various fields, both in theory and practice. Unlike the predictive models, due to the intricate relationship between decisions and the underlying data‐generating processes, the evaluation of decision algorithms cannot directly rely on real data. Hence, a simulator becomes indispensable for appraising decision algorithm effectiveness. In this paper, we aim to leverage assortment decisions, a widely used application in revenue management, to illustrate the utilization of a machine learning‐based simulation. The process can be summarised as: we utilize the modified Transformer‐based choice model, acting as a simulator, to generate a synthetic dataset that mimics consumer purchasing behavior. After training the MNL, DeepFM, and DeepFM‐a models, all of which can swiftly provide assortment decisions in real‐time, we utilize the simulator to evaluate the revenue generated by each assortment prescribed by different choice models. This approach mitigates the challenge of validating decision models that alter real‐world observed data. To show the benefit of such a simulation approach, we have conducted various numerical studies. These studies aim to examine the impact of outside option attractiveness, data size, the number of features, and cardinality. Admittedly, due to the close alignment between the simulator and complex consumer purchase choice datasets, some numerical observations may be challenging to explain. Nevertheless, by employing the simulator, we are able to contrast the differences between the MNL and DeepFM/DeepFM‐a models, shedding light on their respective model misspecifications. [ABSTRACT FROM AUTHOR]

PIWI‐interacting RNAs (piRNAs) are a typical class of small non‐coding RNAs, which are essential for gene regulation, genome stability and so on. Accumulating studies have revealed that piRNAs have significant potential as biomarkers and therapeutic targets for a variety of diseases. However current computational methods face the challenge in effectively capturing piRNA‐disease associations (PDAs) from limited data. In this study, we propose a novel method, MRDPDA, for predicting PDAs based on limited data from multiple sources. Specifically, MRDPDA integrates a deep factorization machine (deepFM) model with regularizations derived from multiple yet limited datasets, utilizing separate Laplacians instead of a simple average similarity network. Moreover, a unified objective function to combine embedding loss about similarities is proposed to ensure that the embedding is suitable for the prediction task. In addition, a balanced benchmark dataset based on piRPheno is constructed and a deep autoencoder is applied for creating reliable negative set from the unlabeled dataset. Compared with three latest methods, MRDPDA achieves the best performance on the pirpheno dataset in terms of the five‐fold cross validation test and independent test set, and case studies further demonstrate the effectiveness of MRDPDA. [ABSTRACT FROM AUTHOR]

This article addresses the evolving landscape of data advertising within network-based new media, seeking to mitigate the accuracy limitations prevalent in traditional film and television advertising evaluations. To overcome these challenges, a novel data-driven nonlinear dynamic neural network planning approach is proposed. Its primary objective is to augment the real-time evaluation precision and accuracy of film and television advertising in the dynamic interactive realm of network media. The methodology primarily revolves around formulating a design model for visual advertising in film and television, customized for the dynamic interactive milieu of network media. Leveraging DeepFM+long short-term memory (LSTM) modules in deep learning neural networks, the article embarks on constructing a comprehensive information statistics and data interest model derived from two public datasets. It further engages in feature engineering for visual advertising, crafting self-learning association rules that guide the data-driven design process and system flow. The article concludes by benchmarking the proposed visual neural network model against other models, using F1 and root mean square error (RMSE) metrics for evaluation. The findings affirm that the proposed model, capable of handling dynamic interactions among images, visual text, and more, excels in capturing nonlinear and feature-mining aspects. It exhibits commendable robustness and generalization capabilities within various contexts. [ABSTRACT FROM AUTHOR]

The quality of OpenStreetMap (OSM) has been widely concerned as a valuable source for monitoring some sustainable development goals (SDG) indicators. Improving its semantic quality is still challenging. As a kind of solution, road type prediction plays an important role. However, most existing algorithms show low accuracy, owing to data sparseness and inaccurate description. To address these problems, we propose a novel OSM road type prediction approach via integrating multiple spatial contexts with DeepFM, named MSC-DeepFM. A deep learning model DeepFM is used for dealing with data sparseness. Moreover, multiple spatial contexts (MSC), including the features of intersecting roads, surrounding buildings, and points of interest (POIs), are distilled to describe multiple types of road more accurately. The MSC combined with geometric features and restricted features are put into DeepFM, in which the low-order and high-order features fully interact. And a multivariate classifier OneVsRest is adopted to predict road types. Experiments on OSM show that the proposed model MSC-DeepFM achieves excellent performance and outperforms some state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Building type prediction is a critical task for urban planning and population estimation. The growing availability of multi-source data presents rich semantic information for building type prediction. However, existing residential building prediction methods have problems with feature extraction and fusion from multi-type data and multi-level interactions between features. To overcome these limitations, we propose a deep learning approach that takes both the internal and external characteristics of buildings into consideration for residential building prediction. The internal features are the shape characteristics of buildings, and the external features include location features and semantic features. The location features include the proximity of the buildings to the nearest road and areas of interest (AOI), and the semantic features are mainly threefold: spatial co-location patterns of points of interest (POI), nighttime light, and land use information of the buildings. A deep learning model, DeepFM, with multi-type features embedded, was deployed to train and predict building types. Comparative and ablation experiments using OpenStreetMap and the nighttime light dataset were carried out. The results showed that our model had significantly higher classification performance compared with other models, and the F1 score of our model was 0.9444. It testified that the external semantic features of the building significantly enhanced the predicted performance. Moreover, our model showed good performance in the transfer learning between different regions. This research not only significantly enhances the accuracy of residential building identification but also offers valuable insights and ideas for related studies. [ABSTRACT FROM AUTHOR]

Recommender systems are an effective way to address the information overload problem. Due to the rapid development of deep learning, recommender systems based on deep learning can efficiently process massive training samples and integrate a variety of additional information, which can alleviate the inherent data sparsity and cold start problems of recommender systems. However, deep learning recommender systems need to collect massive amounts of user data for training models, which poses challenges for data security and user privacy protection. By using users’ local, not centrally collected, data, federated learning can collaboratively train deep learning recommendation models and improve the accuracy of recommendations without compromising user privacy. In this paper, we propose FedDeepFM, a deep recommender system model based on federated learning. First, FedDeepFM follows the standard paradigm of federated learning by uploading each client’s model parameters instead of the original data for model updates, according to the data processing inequality, which ensures that the server cannot collect users’ private information directly but at the same time is able to use the private data held by each client to jointly train a global model with generalization capabilities. Second, to prevent the server from learning users’ private information through indirect inference methods, a pseudo-interaction filling method is used to expand the client data to mask the real user data. The experimental results show that FedDeepFM, the deep recommender system model based on federated learning proposed in this paper, can provide high-quality recommendations while enhancing users’ privacy.

In recent years, deep learning has been applied to the field of recommendation, which can learn complex user interaction features and make better recommendations. However, deep learning only focuses on the interaction of high-order features and neglects the low-order features. The DeepFM model combines the linear FM (Factorization Machines) model and the deep DNN (Deep Neural Network) model to realize the interactive learning of low-order and high-order features, but it does not take into account that user interests will change dynamically with time. When the data sparsity is high, it cannot be effectively recommended. Based on this, an improved DeepFM recommendation algorithm that combines depth feature extraction was proposed, named fDeepFM. Firstly, the word features are transformed into low-dimensional dense vectors through the Embedding layer. Then Doc2Vec is combined to mine item features with context, and the two are stitched together as the input to the FM model and DNN model. Subsequently, user features are input to the GRU (Gated Cyclic Unit) model according to different cycles to mine user features. Finally, the results of the FM model, DNN model, and GRU model are combined by linear stitching as the overall output of the fDeepFM model. Experiments were carried out on Movielens-20M and Amazon data sets. The experimental results showed that MAE, RMSE, F1-score, and AUC on the Movielens-20M data set were optimized by 1.69%, 2.4%, 1.67%, and 2.28%, respectively; On the Amazon dataset, MAE, RMSE, F1-score, and AUC are optimized by 3.2%, 3.86%, 1.63%, and 2.2% respectively compared with DeepFM. [ABSTRACT FROM AUTHOR]

With the increasing popularity of social media represented by Weibo,rumors spread rapidly through social media,which is more likely to cause serious consequences.The problem of automatic rumor detection has attracted widespread attention from academic and industrial circles at home and abroad.We have noticed that more and more users use pictures to post Weibo,not just text.Weibo usually consists of text,images and social context.Therefore,a multi-modal network rumor detection method DCNN based on deep neural network for the text content,image and user attribute information of the accompanying text is proposed.This method consists of a multi-modal feature extractor and a rumor detector.The multi-modal feature extractor is divided into three parts:a text feature extractor based on TextCNN,a picture feature extractor based on VGG-19,and a user social feature extractor based on DeepFM algorithm.These three parts learn feature representations on different modalities of Weibo to form re-parameterized multi-modal features,which are fused as input to the rumor detector classification detection.This algorithm has carried out a large number of experiments on the Weibo data set,and the experimental results show that the recognition accuracy of DCNN algorithm is improved from 78.1% to 80.3%,which verifies the feasibility and effectiveness of DCNN algorithm and feature interaction method for social characteristics.

SUMMARY: As more and more companies and organizations encapsulate and publish their business data or resources to the Internet in the form of APIs, the number of web APIs has grown exponentially. For this reason, it has become challenging to quickly and effectively find web APIs from such a large‐scale web API collection, which meet the requirements of mashup developers. To this end, this article focuses on recommending suitable web APIs to build high‐quality mashups by classifying and integrating content‐oriented service functionality with service invocation prediction. The proposed web API recommendation method for mashup development uses graph attention representation and DeepFM quality prediction. First, it uses the web API composition and shared annotation relationships to construct a web API relationship network. Second, it applies the self‐attention mechanism to compute the attention coefficients of different neighboring nodes in the web API relationship network. So, for a specific web API node, the weighted sum of the importance of its neighboring nodes and features characterizes that web API node. Doing so ensures that the service can be divided more accurately into different functional clusters via high‐quality characterization. Third, for the web APIs in a cluster, the high‐quality representation results are combined with multidimensional quality of service attributes. It employs the DeepFM to model and mine complex interaction relationships between features and subsequently predict and rank the invocation scores of web APIs. Finally, experiments are compared and analyzed on real‐world web API datasets. It can be seen from the results of several groups of comparative experiments that the proposed method outperforms other nine baseline methods on accuracy, recall, F1, DCG, and AUC and achieved a good classification accuracy and recommendation effect. [ABSTRACT FROM AUTHOR]

The traditional financial models used in bond default mainly focus on the analysis and prediction of bonds issued by listed companies, and they lack early warning abilities for a large number of bonds of nonlisted companies. At the same time, there is a great deal of relational data and category data in bond data. It is of great significance for bond default prediction to use these data reasonably, which may bring considerable revenue to companies in the near future. Therefore, this paper uses multisource information from bonds and issuers as well as macroeconomic data to predict bond defaults based on a knowledge graph and deep learning technology. On the basis of constructing a bond knowledge graph, knowledge representation learning technology is used to vectorize the knowledge in the graph, and the extracted vectors are inputted into the deep learning model as features to forecast bond default. The applied model is the deep factorization machine model, and good prediction results are obtained.

In recent years, deep learning has been applied to the field of recommendation, which can learn complex user interaction features and make better recommendations. However, deep learning only focuses on the interaction of high-order features and neglects the low-order features. The DeepFM model combines the linear FM (Factorization Machines) model and the deep DNN (Deep Neural Network) model to realize the interactive learning of low-order and high-order features, but it does not take into account that user interests will change dynamically with time. When the data sparsity is high, it cannot be effectively recommended. Based on this, an improved DeepFM recommendation algorithm that combines depth feature extraction was proposed, named fDeepFM. Firstly, the word features are transformed into low-dimensional dense vectors through the Embedding layer. Then Doc2Vec is combined to mine item features with context, and the two are stitched together as the input to the FM model and DNN model. Subsequently, user features are input to the GRU (Gated Cyclic Unit) model according to different cycles to mine user features. Finally, the results of the FM model, DNN model, and GRU model are combined by linear stitching as the overall output of the fDeepFM model. Experiments were carried out on Movielens-20M and Amazon data sets. The experimental results showed that MAE, RMSE, F1-score, and AUC on the Movielens-20M data set were optimized by 1.69%, 2.4%, 1.67%, and 2.28%, respectively; On the Amazon dataset, MAE, RMSE, F1-score, and AUC are optimized by 3.2%, 3.86%, 1.63%, and 2.2% respectively compared with DeepFM.

In a personalized product recommendation system, when the amount of log data is large or sparse, the accuracy of model recommendation will be greatly affected. To solve this problem, a personalized product recommendation method using deep factorization machine (DeepFM) to analyze user behavior is proposed. Firstly, the K-means clustering algorithm is used to cluster the original log data from the perspective of similarity to reduce the data dimension. Then, through the DeepFM parameter sharing strategy, the relationship between low- and high-order feature combinations is learned from log data, and the click rate prediction model is constructed. Finally, based on the predicted click-through rate, products are recommended to users in sequence and fed back. The area under the curve (AUC) and Logloss of the proposed method are 0.8834 and 0.0253, respectively, on the Criteo dataset, and 0.7836 and 0.0348 on the KDD2012 Cup dataset, respectively. Compared with other newer recommendation methods, the proposed method can achieve better recommendation effect. [ABSTRACT FROM AUTHOR]

Among the inherent problems in recommendation systems are data sparseness and cold starts; the solutions to which lie in the introduction of knowledge graphs to improve the performance of the recommendation systems. The results in previous research, however, suffer from problems such as data compression, information damage, and insufficient learning. Therefore, a DeepFM Graph Convolutional Network (DFM-GCN) model was proposed to alleviate the above issues. The prediction of the click-through rate (CTR) is critical in recommendation systems where the task is to estimate the probability that a user will click on a recommended item. In many recommendation systems, the goal is to maximize the number of clicks so the items returned to a user can be ranked by an estimated CTR. The DFM-GCN model consists of three parts: the left part DeepFM is used to capture the interactive information between the users and items; the deep neural network is used in the middle to model the left and right parts; and the right one obtains a better item representation vector by the GCN. In an effort to verify the validity and precision of the model built in this research, and based on the public datasets ml1m-kg20m and ml1m-kg1m, a performance comparison experiment was designed. It used multiple comparison models and the MKR and FM_MKR algorithms as well as the DFM-GCN algorithm constructed in this paper. Having achieved a state-of-the-art performance, the experimental results of the AUC and f1 values verified by the CTR as well as the accuracy, recall, and f1 values of the top-k showed that the proposed approach was excellent and more effective when compared with different recommendation algorithms.

Among the inherent problems in recommendation systems are data sparseness and cold starts; the solutions to which lie in the introduction of knowledge graphs to improve the performance of the recommendation systems. The results in previous research, however, suffer from problems such as data compression, information damage, and insufficient learning. Therefore, a DeepFM Graph Convolutional Network (DFM-GCN) model was proposed to alleviate the above issues. The prediction of the click-through rate (CTR) is critical in recommendation systems where the task is to estimate the probability that a user will click on a recommended item. In many recommendation systems, the goal is to maximize the number of clicks so the items returned to a user can be ranked by an estimated CTR. The DFM-GCN model consists of three parts: the left part DeepFM is used to capture the interactive information between the users and items; the deep neural network is used in the middle to model the left and right parts; and the right one obtains a better item representation vector by the GCN. In an effort to verify the validity and precision of the model built in this research, and based on the public datasets ml1m-kg20m and ml1m-kg1m, a performance comparison experiment was designed. It used multiple comparison models and the MKR and FM_MKR algorithms as well as the DFM-GCN algorithm constructed in this paper. Having achieved a state-of-the-art performance, the experimental results of the AUC and f1 values verified by the CTR as well as the accuracy, recall, and f1 values of the top-k showed that the proposed approach was excellent and more effective when compared with different recommendation algorithms. [ABSTRACT FROM AUTHOR]