Function Of Decay Time Research Articles

TRGCN: A Prediction Model for Information Diffusion Based on Transformer and Relational Graph Convolutional Network

In order to capture and integrate the structural features and temporal features contained in social graph and diffusion cascade more effectively, an information diffusion prediction model based on the Transformer and Relational Graph Convolutional Network (TRGCN) is proposed. Firstly, a dynamic heterogeneous graph composed of the social network graph and the diffusion cascade graph was constructed, and it was input into the Relational Graph Convolutional Network (RGCN) to extract the structural features of each node. Secondly, the time embedding of each node was reencoded using Bi-directional Long Short-Term Memory (Bi-LSTM). The time decay function was introduced to give different weights to nodes at different time positions, so as to obtain the temporal features of nodes. Finally, structural features and temporal features were input into Transformer and then merged. The spatiotemporal features are obtained for information diffusion prediction. The experimental results on three real datasets of X (formerly known as Twitter), Douban, and Memetracker show that compared with the optimal model in the comparison experiment, the TRGCN model has an average increase of 4.16% in Hits@100 metric and 13.26% in map@100 metric. The validity and rationality of the model are proved.

Personalized Recommendation Systems Powered By Large Language Models: Integrating Semantic Understanding and User Preferences

This study proposes a novel personalized recommendation system leveraging Large Language Models (LLMs) to integrate semantic understanding with user preferences [1]. The system addresses critical challenges in traditional recommendation approaches by harnessing LLMs' advanced natural language processing capabilities. We introduce a framework combining a fine-tuned Roberta semantic analysis model with a multi-modal user preference extraction mechanism.The LLM component undergoes domain adaptation using Masked Language Modeling on a corpus of 112,000 user reviews from the MyAnimeList dataset, followed by task-specific fine-tuning using contrastive learning. User preferences are modeled through a weighted combination of explicit ratings, review sentiment, and implicit feedback, incorporating temporal dynamics through a time-decay function. Experimental results demonstrate significant improvements over state-of-the-art baselines, including Matrix Factorization, Neural Collaborative Filtering, BERT4Rec, and LightGCN. Our LLM-powered system achieves an 8.6%increase in NDCG@10 and a 10.5% improvement in Mean Reciprocal Rank compared to the best-performing baseline. Ablation studies reveal the synergistic effect of integrating LLM-based semantic understanding with user preference modeling. Case studies highlight the system's ability to recommend long-tail items and provide cross-genre suggestions, showcasing its capacity for nuanced content understanding. Scalability analysis indicates that while the LLM-based approach has higher initial computational costs, its performance scales comparably to other deep learning models for larger datasets. This research contributes to the field by demonstrating the effectiveness of LLMs in enhancing recommendation accuracy and diversity. Future work will explore advanced LLM architectures, multi-modal data integration, and techniques to improve computational efficiency and interpretability of recommendations.

Explicable recommendation model based on a time‐assisted knowledge graph and many‐objective optimization algorithm

SummaryExisting research on recommender systems primarily focuses on improving a single objective, such as prediction accuracy, often ignoring other crucial aspects of recommendation performance such as temporal factor, user satisfaction, and acceptance. To solve this problem, we proposed an explicable recommendation model using many‐objective optimization and a time‐assisted knowledge graph, which utilizes user interaction times within the graph to prioritize recommending recently frequently visited items and is further optimized using a many‐objective optimization algorithm. In this model, the temporal weight of user actions at different times is first determined through a time decay function. Additionally, if a user clicks on the same item again, the current action's temporal weight is set to one. This strategy prioritizes recent user actions and frequently visited items, reflecting current interests and preferences better. Next, the created knowledge graph is used to create a list of potential recommendations. Embedding methods obtain the vectors for entities and relations in the path. These vectors, combined with the temporal weight of actions, quantify the explainability of user recommendations. Optimizing the rest of the recommendation performance with many objective algorithms while focusing on the user's recent frequent visits to the item. Finally, the outcomes of the research study indicate that, compared to other explicable recommended methods, our model, considering temporal factor, improved average accuracy by 11%, diversity by 1%, and explainability by 21% in the Useraction1 data set. Results in other data sets also indicate that the proposed model maintains accuracy, diversity, and novelty while enhancing explainability.

Transformer-based modeling of abnormal driving events for freeway crash risk evaluation

A crash risk evaluation model aims to estimate crash occurrence possibility by establishing the relationships between traffic flow status and crash occurrence. Based upon which, Proactive Traffic Safety Management (PTSM) systems have been developed and implemented. The current crash risk evaluation models relied on high dense traffic detectors, which limited the applications of PTSM to infrastructures with enough sensing devices. To address such application limitation issue, this study employed the widespread abnormal driving event information that is generated by emerging driving monitoring and vehicle connection techniques to develop the crash risk evaluation model. Specifically, to characterize abnormal driving events, a six-tuple embedding method was proposed to store their space, time and kinetics features. Given their irregular and discrete distributions on roadways, a Transformer model with self-attention mechanism was proposed to extract the spatial distribution characteristics. In addition, a time-decay function was integrated to fit the temporal impacts of abnormal driving events on crash risk. Empirical data from a freeway in China were utilized for the analyses. The results showed that abnormal driving events with lower speed, larger acceleration and duration are more likely to cause crashes. The accumulation of multiple events in the time period of less than 3 min would lead to a sharp increase of crash risk. Besides, compared to the average metrics of the widely adopted Convolutional Neural Network (CNN), XGBoost, and logistic regression models, the proposed model achieved higher accuracy (0.841) and AUC (0.777), with average improvement of 2.5 % and 9.1 % respectively.

A recommendation model for e-commerce platforms oriented to explicit information compensation and hidden information mining

In an e-commerce platform recommendation system, accurately mining users’ preferences and hidden information has always been an important initiative. In this study, based on the potential preferences and hidden characteristics of users, we propose a recommendation model for e-commerce platforms oriented to explicit information compensation and hidden information mining. First, considering the advantages of the VGAE (variational Graph Auto-Encoder) in reconstructing data samples, we propose an EI-VGAE (Explicit Information-Variational Graph Auto-Encoder) to compensate for the lack of explicit feedback. Second, to address the problem that hidden information in e-commerce platforms is difficult to mine, an xDeepFM (eXtremely Deep Factorization Machine) is used to jointly learn higher-order interaction features explicitly and implicitly. A time decay function is also introduced to reduce the impact of higher-order interaction features on the representation of hidden information. Finally, considering the advantages of the attention mechanism for obtaining implicit relational attention and the quantification of the user’s degree of preference in local representation, a GAT (Graph Attention Network)-based recommendation model for the e-commerce platform EI-GAT (Explicit and Implicit Graph Attention Network) is proposed. Our experiments show that the proposed method effectively alleviates the problem of low recommendation accuracy owing to insufficient explicit feedback from consumers and further improves recommendation accuracy by quantifying user preferences.

Time-Aware Based Recommendation System using Gower’s Coefficients: Enhancing Personalized Recommendation

To give users personalised recommendations, the collaborative filtering technique is widely used. In this technique, similarity computation plays a crucial role. Once the similarities among users or items are determined, the system can predict how a user might rate or interact with items that they have not yet encountered. The traditional approaches or measures mainly considered the user-item historical ratings to compute the similarity, while user preferences may change with time. Considering this, the objective of this research is to create an efficient recommendation system that utilized the temporal data. For this, few time decay functions, i.e., exponential, linear, logistic, and power applied to the ratings to give more weightage to the most recent ratings. Since, Gower’s coefficient is suitable for handling the missing data, it is applied to calculate the similarity and compared its results with popular traditional similarity measures i.e., Cosine and Pearson correlation coefficient. Experimental findings on the ML-100k dataset in terms of Root Mean Squared Error (RMSE) and Mean Average Error (MAE) demonstrate that performance of RS improved when we applied a power function on ratings. With comparison to most recent methods PIP, NHSM, RJaccard etc., proposed approach gives almost 10% better results in MAE and 5% better results in RMSE comparison.

A Novel Numerical Modeling Chinese Population Evolution

In this paper, we build a very accurate mortality model with a constant time decay function and the optimal goal of minimizing the total number of deaths, then solve it by a numerical approach. China’s vast territory, large population, complex flow of people, and the concept of returning to the hometown greatly affect the accuracy of survey population mortality. Furthermore, the instability of the solution of the Lee-Carter mortality model makes it difficult to predict the future population mortality in China. Based on the population data published in China from 1990 to 2000 with corrections, the mortality results from 2001 to 2019 are predicted by using the new model. Compared with the population and deaths in the National Statistical Bulletin, the average death rate error is less than 0.64%, the maximum annual error is less than 6.11%. According to this result, the total population and mortality results from 2001 to 2020 are predicted. Most of the prediction error rates of annual total population is 1 ‰, and the maximum error rate is less than 3.4 ‰. The accuracy of the model for Chinese population prediction is around 10 times higher than any other current models.

An electronic medical record access control model based on intuitionistic fuzzy trust

Electronic Medical Records (EMR), as the core data of medical big data, has improved the efficiency of medical services in the process of sharing and use, but a series of data privacy leakage problems have emerged. Access control technology can ensure users' legitimate and appropriate access to data resources, but traditional access control has many limitations such as static and coarse-grained, which is difficult to meet in the dynamic and complex healthcare environment. We create a model to solve the privacy leakage problem of healthcare big data. The model utilizes the Latent Delicacy Allocation (LDA) topic model to quickly search documents and extract the topic probability distributions between the target patient and other patients' electronic medical records, which improves the efficiency of medical record document analysis; Calculating physician access similarity, we quantify physician access behavior to more accurately describe physician access behavior in the form of data. We combined the intuitionistic fuzzy theory with the trust evaluation method to create a novel trust evaluation method-- Intuitionistic Fuzzy Trust Evaluation Method, which can reflect the uncertainty characteristics that exist in the physician's diagnostic and treatment process when quantifying the trust of the access by using this method, and it is more flexible and practical, which enhances the accuracy of the quantification of the trust. Improving the traditional trust aggregation method by adding a time window and time decay function, the optimized method improves the accuracy of trust aggregation, and the experimental results are more in line with the actual law. We created a new intuitionistic fuzzy trust access control model using a relative entropy-based intuitionistic fuzzy clustering algorithm and a reward and punishment mechanism to achieve adaptive and dynamic access control for physician access behavior. Experiments show that the access control model based on intuitionistic fuzzy trust proposed in this article achieves 95% correctness in identifying user access behaviors and 94.29% correctness in identifying excessive user access behaviors. Experiments have proved that this model has obvious advantages over other models in the process of privacy protection of electronic medical record systems and has certain control effects.

Technical note: Excel spreadsheet calculation of the Henssge equation as an aid to estimating postmortem interval

In forensic cases for which the time of death is unknown, several methods are used to estimate the postmortem interval. The quotient (Q) defined as the difference between the rectal and ambient temperature (Tr − Ta) divided by the initial difference (T0 − Ta) represents the progress of postmortem cooling: Q = (Tr − Ta)/(T0 − Ta), (1 ≥ Q ≥ 0). Henssge was able to show that with the body weight and its empirical corrective factor, Q can be reasonably predicted as a double exponential decay function of time (Qp(t)). On the other hand, actual Q is determined as Qd by measuring Tr and Ta under an assumption of T0 = 37.2 °C. Then, the t value at which Qp(t) is equal to Qd (Qd=Qp(t)) would be a good estimate of the postmortem interval (the Henssge equation). Since the equation cannot be solved analytically, it has been solved using a pair of nomograms devised by Henssge. With greater access to computers and spreadsheet software, computational methods based on the input of actual parameters of the case can be more easily utilized. In this technical note, we describe two types of Excel spreadsheets to solve the equation numerically. In one type, a fairly accurate solution was obtained by iteration using an add-in program Solver. In the other type (forward calculation), a series of Qp(t) was generated at a time interval of 0.05 h and the t value at which Qp(t) was nearest to Qd was selected as an approximate solution using a built-in function, XLOOKUP. Alternatively, a series of absolute values of the difference between Qd and Qp(t) (|Dq(t)| = |Qd − Qp(t)|) was generated with time interval 0.1 h and the t value that produces the minimum |Dq(t)| was selected. These Excel spreadsheets are available as Supplementary Files.

Event-based incremental recommendation via factors mixed Hawkes process

Incremental recommendation systems have garnered significant research interest since they ideally adapt to users' ongoing events (such as clicking, browsing, and reviewing) and recommend items without retaining the model. Many methods have tracked the event generation sequences for incremental recommendation. However, most existing models treat the event as a static snapshot or a black box, ignoring the underlying factors that may trigger the event generation. Underlying such inner factors can help RS reasonably and foreseeingly evaluate the potential items for the user next time. Along this vein, we propose the Factors Mixed Hawkes Process (FMHP) for event-based incremental recommendations. First, we extend each event to a notion of factor-driven event sequence. Next, we consider three factors that may influence the occurrence of an event: intrinsic intensity, external intensity, and historical intensity. An intrinsic intensity function, multi-type temporal attention, and a hybrid time decay function are incorporated in FMHP to evaluate the intrinsic, external, and historical intensity, respectively. In addition, an incremental updating strategy is implemented in FMHP, continuously updating event intensity as new events occur. We conduct extensive experiments on four public datasets (e.g., Amazon Beauty, LastFM, Movielens, and Amazon Book). Compared with state-of-the-art incremental recommendation methods, our proposed FMHP model achieves superior performance, with up to 9.77%, 9.35%, 9.32%, 10.10% w.r.t. HR, 8.86%, 10.26%, 9.81%, 9.38% w.r.t. NDCG, and 9.64%, 9.32%, 8.97%,9.85% w.r.t. Recall in Beauty, LastFM, MovieLens, and Book, respectively. Besides, the case study shows that the three factors in our proposed FMHP method play a vital role in triggering event generation.

DOIDS: An Intrusion Detection Scheme Based on DBSCAN for Opportunistic Routing in Underwater Wireless Sensor Networks.

In Underwater Wireless Sensor Networks (UWSNs), data should be transmitted to data centers reliably and efficiently. However, due to the harsh channel conditions, reliable data transmission is a challenge for large-scale UWSNs. Thus, opportunistic routing (OR) protocols with high reliability, strong robustness, low end-to-end delay, and high energy efficiency are widely applied. However, OR in UWSNs is vulnerable to routing attacks. For example, sinkhole attack nodes can attract traffic from surrounding nodes by forging information such as the distance to the sink node. In order to reduce the negative impact of malicious nodes on data transmission, we propose an intrusion detection scheme (IDS) based on the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering algorithm for OR (DOIDS) in this paper. DOIDS is based on small-sample IDS and is suitable for UWSNs with sparse node deployment. In DOIDS, the local monitoring mechanism is adopted. Every node in the network running DOIDS can select the trusted next hop. Firstly, according to the behavior characteristics of common routing attack nodes and unreliable underwater acoustic channel characteristics, DOIDS selected the energy consumption, forwarding, and link quality information of candidate nodes as the detection feature values. Then, the collected feature information is used to detect potential abnormal nodes through the DBSCAN clustering algorithm. Finally, a decision function is defined according to the time decay function to reduce the false detection rate of DOIDS. It makes a final judgment on whether the potential abnormal node is malicious. The simulation results show that the algorithm can effectively improve the detection accuracy rate (3% to 15% for different scenarios) and reduce the false positive rate, respectively.

Performance Evaluation of Time-based Recommendation System in Collaborative Filtering Technique

The Collaborative Filtering (CF) technique is the most common neighbourhood-based recommendation strategy, that provides personalized recommendation to a user for the items using a similarity measure. Hence, the selection of the appropriate similarity measure becomes crucial in the CF based recommendation system. The traditional similarity measures merely focus only on the historical ratings provided by the users to compute the similarity, completely ignoring the fact that preferences change over a period of time. Considering this, the paper aims to develop an effective Recommendation System that uses temporal information to capture the changes in the preferences over a period of time. For this, the existing exponential and power time decay functions are integrated with Cosine, Pearson Correlation, and Gower's similarity measures to compute similarity. The similarity is computed at the similarity computation and prediction levels of recommendation processes. Experimental findings in terms of MAE and RMSE on the MovieLens-100k demonstrate that performance of Gower's coefficient is better when applied with the exponential function at the similarity computation level of the recommendation process.

Contractor Recommendation Model Using Credit Networking and Collaborative Filtering

The credit of contractors in the construction market directly affects the cooperative intentions of owners. Although previous scholars have attempted to use credit to select appropriate contractors, they have rarely considered the trust relationship between decision-making and former owners. This work introduces and verifies a credit network recommendation model based on a collaborative filtering algorithm. The contractor’s credit established based on this model serves as a viable method for owners to select efficient contractors. The application of the model includes relevant information collection, neighbor set formation, contractor’s credit evaluation, and recommendation list formation, among which the neighbor set of the owner is used to calculate the comprehensive trust degree of the decision-making owner to the former owner. A time decay function is adopted to correct the difference in the trust relationship between an owner and a contractor introduced over time. To verify the feasibility of this model, an actual scenario was simulated, and the results obtained via simulations were compared and found to be consistent. Thus, a contractor with a high credit can be recommended to the decision-making owner. This approach is crucial for promoting contractors’ credit and conducive to the healthy development of the construction market.

User Identity Linkage across Social Networks with the Enhancement of Knowledge Graph and Time Decay Function.

Users participate in multiple social networks for different services. User identity linkage aims to predict whether users across different social networks refer to the same person, and it has received significant attention for downstream tasks such as recommendation and user profiling. Recently, researchers proposed measuring the relevance of user-generated content to predict identity linkages of users. However, there are two challenging problems with existing content-based methods: first, barely considering the word similarities of texts is insufficient where the semantical correlations of named entities in the texts are ignored; second, most methods use time discretization technology, where the texts are divided into different time slices, resulting in failure of relevance modeling. To address these issues, we propose a user identity linkage model with the enhancement of a knowledge graph and continuous time decay functions that are designed for mitigating the influence of time discretization. Apart from modeling the correlations of the words, we extract the named entities in the texts and link them into the knowledge graph to capture the correlations of named entities. The semantics of texts are enhanced through the external knowledge of the named entities in the knowledge graph, and the similarity discrimination of the texts is also improved. Furthermore, we propose continuous time decay functions to capture the closeness of the posting time of texts instead of time discretization to avoid the matching error of texts. We conduct experiments on two real public datasets, and the experimental results show that the proposed method outperforms state-of-the-art methods.

A Dynamic Heterogeneous Information Network Embedding Method Based on Meta-Path and Improved Rotate Model

Aiming at the current situation of network embedding research focusing on dynamic homogeneous network embedding and static heterogeneous information network embedding but lack of dynamic information utilization, this paper proposes a dynamic heterogeneous information network embedding method based on the meta-path and improved Rotate model; this method first uses meta-paths to model the semantic relationships involved in the heterogeneous information network, then uses GCNs to get local node embedding, and finally uses meta-path-level aggression mechanisms to aggregate local representations of nodes, which can solve the heterogeneous information utilization issues. In addition, a temporal processing component based on a time decay function is designed, which can effectively handle temporal information. The experimental results on two real datasets show that the method has good performance in networks with different characteristics. Compared to current mainstream methods, the accuracy of downstream clustering and node classification tasks can be improved by 0.5~41.8%, which significantly improves the quality of embedding, and it also has a shorter running time than most comparison algorithms.

A Hybrid Approach to Trust Node Assessment and Management for VANETs Cooperative Data Communication: Historical Interaction Perspective

Vehicular ad hoc networks (VANETs) provide self-organized wireless multihop transmission, where nodes cooperate with each other to support data communication. However, malicious nodes may intercept or discard data packets, which might interfere with the transmission process and cause privacy leakage. We consider historical interaction data of nodes as an important factor of trust. Thus, this paper focuses on the trust node management of VANETs, which aims to quantify node credibility as an assessment method and avoid assigning malicious nodes. First, the integrated trust of each node is proposed, which consists of the direct trust and the recommended trust. The former is dynamically computed by historical interaction records and Bayesian inference considering penalty factors. The latter defines trust by third-party nodes and their reputation. Second, the process of trust calculation and data communication calls for timeliness. Therefore, we introduce a time sliding window and time decay function to ensure that the latest interaction information has a higher weight. We can sensitively identify malicious nodes and make quick responses. Finally, the experimental results demonstrate that our proposed method outperforms bassline methods, especially with respect to the packet delivery ratio and security.

Recommendation Model Based on Dynamic Interest Group Identification and Data Compensation

With the increasing network service content, innovative methods are required for developing optimized network service for e-commerce companies. Accordingly, this study focuses on designing a framework containing personalization, interest group identification, and recommendation mechanisms. The primary contribution of this paper is to propose a recommendation model based on data compensation and dynamic user interest grouping. First, to address the problem of sparse user rating data, homeostasis compensation is performed on native data to more realistically restore the preference relationship between users and items by introducing the advantages of generative adversarial network in learning data distribution and enhancing data samples. Second, to address the problem of user interest generalization, information entropy is introduced to measure the user interest feature space. In addition, the time window marking method is used to further quantify the users’ dynamic interest group around the users’ interest drift. Finally, considering tensor decomposition characteristics in data dimension transformation and data compression, a score prediction model based on the “user-item-interest group” tensor decomposition is constructed. Simultaneously, a time decay function is introduced in the construction of the tensor to dynamically fit the user behavior and further improve prediction accuracy. Experiments show that the proposed framework can effectively improve the recommendation accuracies resulting from both sparse scoring data and dynamic user interest division.

A social rumor and anti-rumor game diffusion model based on sparse representation and tensor completion

In view of the high dimensionality and complexity of the rumor diffusion space in social networks, and considering the game confrontation in the process of rumor diffusion, this paper proposes a rumor and anti-rumor game diffusion model based on sparse representation and tensor completion. First, taking advantage of the sparse representation’s ability to represent all of the original samples with as few atoms as possible to enable low-rank vectorization of the feature space. Second, in view of the fact that a tensor complement can recover lost data with high precision, a tensor complement with a time-decay function is used to complement the dynamic behavior data of users during the rumor process. Finally, considering the game relationship between rumors and anti-rumors during diffusion, evolutionary game theory is introduced. In addition, this paper proposes a cooperation-and-competition graph convolutional network (CC-GCN)-based model for predicting user behavior. Experiments demonstrate that the CC-GCN model not only improves the prediction accuracy of user behavior but also clearly reflects the game relationship between rumors and anti-rumors in the diffusion space.

Dynamic Influence Ranking Algorithm Based on Musicians’ Social and Personal Information Network

Social influence analysis is a very popular research direction. This article analyzes the social network of musicians and the many influencing factors when musicians create music to rank the influence of musicians. In order to achieve the practical purpose of the model making accurate predictions in the broad music market, the algorithm adopts a macromodel and considers the social network topology network. The article adds the time decay function and the weight of genre influence to the traditional PageRank algorithm, and thus, the MRGT (Musician Ranking based on Genre and Time) algorithm appears. Considering the timeliness of social networks and the continuous development of music, we realized the importance of evolving MRGT into a dynamic social network. Therefore, we adopted audio data analysis technology and used Gaussian distance to classify and study the evolution of music properties at different times and different genres and finally formed the dynamic influence ranking algorithm based on musicians’ social and personal information networks. As a macromodel heuristic algorithm, our model is explanatory, can handle batch data and can avoid unfavorable factors, so as to provide fast speed and improved accuracy. The network can obtain an era indicator DMI (Dynamic Music Influence) that measures the degree of music revolution. DMI is the indicator we provide for music companies to invest in musicians.

Literary work education model based on intelligent machine learning and reader scoring criteria factors

In order to improve the effect of literary works education, this paper combines intelligent machine learning and reader scoring criteria factors to construct an intelligent education model, and proposes a collaborative filtering recommendation algorithm based on item proportion factors and time decay. When calculating the user similarity, this paper adds the scale factor of the intersection of common scoring items to all the scoring items, and considers the non-intersection part of the user scoring items. Secondly, when predicting the project score, this paper adds a time decay function, combines the forgetting curve law to modify the score prediction method, and combines the actual needs to construct the basic framework of the education model. In addition, this paper designs experiments to verify the performance of the literary work education model constructed in this paper. The research results show that the literary work education model constructed in this paper based on intelligent machine learning and reader rating criteria factors has a certain role in promoting the effect of literary education.