Extracting more information from feature interactions is essential to improve click-through rate (CTR) prediction accuracy. Although deep learning technology can help capture high-order feature interactions, the combination of features lacks interpretability. In this paper, we propose a multi-semantic feature interaction learning network (MeFiNet), which utilizes convolution operations to map feature interactions to multi-semantic spaces to improve their expressive ability and uses an improved Squeeze & Excitation method based on SENet to learn the importance of these interactions in different semantic spaces. The Squeeze operation helps to obtain the global importance distribution of semantic spaces, and the Excitation operation helps to dynamically re-assign the weights of semantic features so that both semantic diversity and feature diversity are considered in the model. The generated multi-semantic feature interactions are concatenated with the original feature embeddings and input into a deep learning network. Experiments on three public datasets demonstrate the effectiveness of the proposed model. Compared with state-of-the-art methods, the model achieves excellent performance (+0.18% in AUC and -0.34% in LogLoss VS DeepFM; +0.19% in AUC and -0.33% in LogLoss VS FiBiNet).

Behavior analysis has been used widely in antifraud transactions. However, existing methods of behavior analysis mainly focus on behavior patterns and do not fully consider the behavior psychology of users in the transaction process, which affects the precise recognition of fraudulent behaviors. It is difficult to recognize fraudulent transactions precisely how to describe the user’s behavioral psychology and integrate the behavioral psychology into the transaction behavior. Thus, this article first proposes the transaction character model based on the user’s cautiousness to reflect the user’s behavioral psychology. This model is built from the user’s historical normal interaction behavior data. Then, a user behavior benchmark is established to reflect the user’s behavior pattern from the user’s historical normal transaction behavior data. To integrate the user’s transaction character and user behavior, the mapping relationship model is built by using the least-squares generalized inverse method. This model is the core of the fraudulent behavior recognition method with transaction characters. Experiments in fraud detection scenarios show that the new method improved the average recognition performance of four fraud detection indicators (recall rate, precision rate, accuracy rate, and F1 value) by 23%. The method also shows that individual psychological character has a great influence on user behavior.

Cloud-edge collaboration is an emerging approach that combines cloud computing and edge computing. This combination holds great potential for enhancing system efficiency and reliability, as well as mitigating processing costs and latency. Existing research focuses on task scheduling and service deployment in cloud or edge environments, aiming to achieve different optimization goals. However, traditional deployment methods are no longer suitable for the current used cloud-edge collaboration model, which does not consider the heterogeneity of the cloud-edge, the constrained relationship between cost and latency, and the distribution of user access. To this end, this paper proposes a service deployment scheme to minimize latency in a heterogeneous cloud-edge environment. The scheme first introduces a cloud-edge pre-distribution mechanism (ACEP) to reasonably divide the services among the cloud or edge servers, and then employs a global heuristic latency optimization algorithm (GLOA) and a cost allocation strategy (CACS) to find the best deployment strategy. The effectiveness of our proposed method is evaluated by comparing it with existing algorithms, which shows its superiority in terms of latency.

Multi-behavior recommendation has gained significant attention in recent years for its ability to outperform singlebehavior models.Current research related to multi-behavior models leaves room for improvement in the following two areas.First, the noise carried by individual behaviors and the additional noise generated during behavior processing is often overlooked, and these can ultimately degrade recommendation performance.Second, the specific time period of behavioral interactions and the frequency of interactions within that time period are also not taken into account.To address the above limitations, we propose a multibehavior recommendation model integrating dynamic preferences (MB-DP) that captures dynamic interests while smoothing and denoising multi-behavior information.MB-DP extracts low and high-order semantics from various behaviors and unifies the measurements to generate interaction predictions.Additionally, it analyzes the interaction time and frequency of each behavior using gated recurrent units to capture the dynamic preferences of users and improve the prediction values.Extensive experimental results on two real-world datasets show that MB-DP significantly improves recommendation performance compared to the state-ofthe-art baselines.

The sampling method for real-time and high-speed changing streaming data is prone to lose the value and information of a large amount of discrete data, and it is not easy to make an efficient and accurate streaming data valuation.The SDSLA (Streaming Data Drilling Sampling Method Under Limited Access) sampling method based on mineral drilling exploration can streaming data valuation containing many discrete data in real-time, but when the range of discrete data in streaming data is irregular, it has low sampling accuracy for discrete data.Based on the SDSLA algorithm, we propose a dynamic drilling sampling method SDDS (Streaming Data Dynamic Drilling Sampling).This method takes well as the analysis unit dynamically changes the size and position of the well, and accurately predicts the position and range of discrete data.A new model SDVEM (Streaming Data Value Evaluation Model), is further proposed for data valuation, which evaluates the sample set from discrete, centralized, and overall dimensions.Experiments show that the method proposed in the paper uses neural network training and testing with a small sampling rate to obtain accuracy, recall, and F1 scores above 90%, which is higher than that of the SDSLA algorithm.In summary, the SDDS sampling method is beneficial to the training neural network models and evaluating the value characteristics of streaming data, which has essential research significance in big data valuation.

Multi-behavior recommendation has gained significant attention in recent years for its ability to outperform single-behavior models. Current research related to multi-behavior models leaves room for improvement in the following two areas. First, the noise carried by individual behaviors and the additional noise generated during behavior processing is often overlooked, and these can ultimately degrade recommendation performance. Second, the specific time period of behavioral interactions and the frequency of interactions within that time period are also not taken into account. To address the above limitations, we propose a multi-behavior recommendation model integrating dynamic preferences (MB-DP) that captures dynamic interests while smoothing and denoising multi-behavior information. MB-DP extracts low and high-order semantics from various behaviors and unifies the measurements to generate interaction predictions. Additionally, it analyzes the interaction time and frequency of each behavior using gated recurrent units to capture the dynamic preferences of users and improve the prediction values. Extensive experimental results on two real-world datasets show that MB-DP significantly improves recommendation performance compared to the state-of-the-art baselines.

The big data sampling method for real-time and high-speed streaming data is prone to lose the value and information of a large amount of discrete data, and it is not easy to make an efficient and accurate evaluation of the value characteristics of streaming data. The SDSLA sampling method based on mineral drilling exploration can evaluate the valuable information of streaming data containing many discrete data in real-time, but when the range of discrete data is irregular, it has low sampling accuracy for discrete data. Based on the SDSLA algorithm, we propose a dynamic drilling sampling method SDDS, which takes well as the analysis unit, dynamically changes the size and position of the well, and accurately locates the position and range of discrete data. A new model SDVEM is further proposed for data valuation, which evaluates the sample set from discrete, centralized, and overall dimensions. Experiments show that compared with the SDSLA algorithm, the sample sampled by the SDDS algorithm has higher evaluation accuracy, and the probability distribution of the sample is closer to the original streaming data, with the AOCV indicator being nearly 10% higher. In addition, the SDDS algorithm can achieve over 90% accuracy, recall, and F1 score for training and testing neural networks with small sampling rates, all of which are higher than the SDSLA algorithm. In summary, the SDDS algorithm not only accurately evaluates the value characteristics of streaming data but also facilitates the training of neural network models, which has important research significance in big data estimation.

As an increasingly mature business model, crowdsourcing, especially spatial crowdsourcing, has played an important role in data collection, disaster response, urban planning and other fields. However, the rapid growth of user scale and massive data collected inevitably brings serious challenges to computing and storage resources. The emergence of cloud computing provides an opportunity to handle such challenges. Its nearly unlimited resource provision capability can provide reliable services for different crowdsourcing applications. Nevertheless, considering the risks of privacy leakage and vendor lock-in using only a single cloud, as well as the additional restrictions caused by the wide geographical distribution of data and associations among workers, the use of multi-cloud seems to be a better choice. In this article, we define a problem to find an effective data placement scheme for spatial crowdsourcing in multi-cloud environment to achieve the cost-effectiveness and minimal latency. We take full account of the interval pricing strategy. Then we analyze the geographical distribution characteristics of data centers through a clustering algorithm, and propose an effective data initialization strategy. Finally, we use a genetic algorithm to further optimize the results. Through experiments on real-world data from cloud providers, the efficiency and effectiveness of our proposed method is verified. Compared with some existing algorithms, the proposed method can significantly reduce the system cost and latency, among which the cost reduction is up to 150 times and the latency reduction is up to twice.

With the development of cloud computing, the cloud market is becoming more and more complicated. There are many cloud providers and different cloud instance types, which brings users some confusion when they select cloud instance types. In order to solve the cloud instance type selection problem in multi-cloud environment, a Cloud Instance Type Selection Algorithm based on Genetic Algorithm (CITSA-GA) is proposed. CITSA-GA mainly includes two-dimensional encoding with the constraint between adjacent genes, selection operation adopting the elite retention strategy and the roulette strategy, crossover operation using the first fit strategy, and mutation operation with mutation bounds. We perform some experiments to prove the effectiveness of the proposed CITSA-GA.

With the rapid development of Internet finance, the volume of online transactions increases gradually, but the risk of exposure is increasing, and fraud is emerging. Because of the characteristics of online transaction, such as large volume, high frequency and fast update speed. In addition, online transaction data has the problems of unbalanced positive and negative sample and sparse timing of transaction data. Most of the existing methods to solve data imbalance are sampled, but this method will change the dataset’s distribution, which is not conducive to improving the generalization ability of the model. There are some timing characteristics of online transaction data, and the common fraud detection model does not take the problem into account in the design of the model. Based on the problems, this paper puts forward the siamese neural network structure based on CNN and LSTM, uses the siamese neural network structure to solve the problem of sample imbalance in online transaction and uses the LSTM structure to make model memory user's transaction information, in order to better detect the fraudulent transaction. The model presented in this paper is verified in real B2C transaction data, and its precision and recall reach about 95% and 96%, respectively.

In large-scale network service systems, the phenomenon of instantaneous gathering of a large number of users can cause system abnormality, whenever the load imposed by the user behaviors does not match the system load. This paper proposes a behavior reconstruction model for large-scale network service systems integrated with Petri net reconstruction methodology, for the purpose of achieving load balancing in the system under increasing number of users. Based on the features of the user interaction behavior sequence, the behavioral load balancing model defines a user behavior membership function. Then, a random fuzzy Petri net with delay is presented to control the user behavior reconstruction. Experiments conducted by considering various changes in the number of user behaviors and their distribution in unit time demonstrate that the proposed methodology can effectively trigger the reconstructed model to balance the system load when the system load exceeds the defined warning point.

Using wireless mobile terminals has become the mainstream of Internet transactions, which can verify the identity of users by passwords, fingerprints, sounds, and images. However, once these identity data are stolen, traditional information security methods will not avoid online transaction fraud. The existing convolutional neural network model for fraud detection needs to generate many derivative features. This paper proposes a fraud detection model based on the convolutional neural network in the field of online transactions, which constructs an input feature sequencing layer that implements the reorganization of raw transaction features to form different convolutional patterns. Its significance is that different feature combinations entering the convolution kernel will produce different derivative features. The advantage of this model lies in taking low dimensional and nonderivative online transaction data as the input. The whole network consists of a feature sequencing layer, four convolutional layers and pooling layers, and a fully connected layer. Verifying with online transaction data from a commercial bank, the experimental results show that the model achieves excellent fraud detection performance without derivative features. And its precision can be stabilized at around 91% and recall can be stabilized at around 94%, which increased by 26% and 2%, respectively, comparing with the existing CNN for fraud detection.

Refinement operation of workflow nets is provided in this paper for modeling and analyzing integrated workflow. Structure, dynamic properties and behavior expression of refined workflow net are also discussed. These works proves that refinement operation with step-by-step refinement of transitions could realize hierarchical modeling of workflow as well as composite modeling of workflow integration Furthermore, the refinement operation can reduce complexity of model analysis. In fact the reliable refined nets satisfy soundness, and dynamic behavior of refined Petri nets was consistent with of original nets and subnets. Therefore the properties analysis and verification of refined workflow nets can be realized by properties of subnets using refinement operation. Moreover, the research results are successfully applied to designing, modeling and verification of layered workflows and their integration in Shanghai Traffic Information Service Application Grid (STISAG).