Search

Air Traffic Flow Management regulations consist of issuing pre-departure ground delays to deal with demand-capacity imbalances in the airspace, smoothing out the demand at congested infrastructures. This results in discrepancies between the initially planned and executed (regulated) flights, negatively impacting operational efficiency and economic performance. To mitigate downstream effects, airspace users need to assess the severity and impact of these regulations when defining their operational plan. It is crucial to anticipate these potential network issues and plan for mitigation measures during the pre-tactical phase (the day before operations (D−1)). This article presents four independent machine learning models to estimate the impact of Air Traffic Flow Management regulations for individual flights during this pre-tactical phase. An integrated view is proposed to combine the results of the different models, displaying the appropriate level of information. Two models are designed to provide information about the probability of a given flight being affected by Air Traffic Flow Management regulations and the reason for this regulation (airport or airspace congestion). These models have reported accuracy levels between 82% and 87%. The remaining models estimate the impact of the delay (amount of Air Traffic Flow Management delay issued to the flight if regulated), with a Mean Absolute Error of 9.5 min when predicting this amount of delay. The SHapely Additive exPlanations analysis is used to identify the most important factors for detecting Air Traffic Flow Management regulations for individual flights during the pre-tactical phase from a data-driven perspective. An integrated view is proposed to create a system which combines the results of the four different models, displaying the appropriate level of information and avoiding an overload of information. Finally, the results are compared against models considering only static or post-operational data. • Machine learning to predict pre-tactical air traffic flow management per flight. • Expert system to support airline's operation planning and monitoring. • Use of data available at pre-departure ensuring operational applicability. • Differentiation between static and dynamic features. • High accuracy with operational relevant information provided to end users. [ABSTRACT FROM AUTHOR]

With the exponential growth of data volume, information overload has become a problem for users, and recommender systems were employed to solve this problem. The purpose of the recommender system is to mine the information of interest to users from massive amounts of data. However, some problems have not been well addressed in recommender systems, e.g., the sparse interaction data between users and items and the cold-start problems when making recommendations to new users. In recent years, incorporating knowledge graphs as side information to recommender systems by knowledge graph embedding techniques has attracted considerable interest, because the rich information contained in the knowledge graph can effectively solve the above problems. This paper provides a systematic review of recommender systems based on knowledge graph embedding in terms of methods and applications. Specifically, some basic notions of recommender systems and knowledge graphs are first briefly introduced, followed by a detailed description of how existing methods associate knowledge graph embedding and recommender systems. In addition, a series of related recommendation application scenarios are summarized along with information and statistics on related datasets. [ABSTRACT FROM AUTHOR]

The recommender systems are one of the most effective big data tools for solving the information overload problem, but data sparsity greatly affects its performance. However, most of the existing graph-based contrastive learning methods perturb the original graph structure for data augmentation, which may lead to semantic loss and subsequently degrade the recommendation performance. Meanwhile, studies have shown that improving the uniformity of data distribution is more effective than data augmentation using graph perturbation. In this paper, to enhance the uniformity of data distribution, we propose D istribution- A ware H ybrid N oise augmentation in graph contrastive learning for Rec ommendation (DAHNRec). Specifically, we design distribution-aware hybrid noise augmentation (DAHN) method to generate noise suitable for the users and items, instead of simply utilizing random noise following uniform distributions. By applying DAHN, the model can learn the appropriate noise distribution during the training process to enhance the uniformity of the embedding space. Additionally, we propose Balanced Bayesian Personalized Ranking (BBPR) as the loss function for recommendation tasks to improve difference characterizing between positive and negative samples. This improves the ranking tasks' performance. Furthermore, we conduct extensive experiments based on public benchmark datasets. The results indicate that DAHNRec outperforms several existing methods. Furthermore, DAHNRec is a general recommendation model based on user–item interactions, which can be used to improve the recommendation performance in various real-world scenarios, such as e-commerce platforms and social media. • A distribution-aware hybrid noise augmentation boosts contrastive tasks. • New loss function enhances recommendation performance. • Ensuring data distribution uniformity is beneficial for graph-based contrastive recommendation. [ABSTRACT FROM AUTHOR]

The complexity associated with groups of tourists led to the emergence of Group Recommender Systems (GRS) for tourism. But if generating recommendations for small groups is a complex task, to provide them to large and occasional groups is even more. This complexity is especially due to the group's heterogeinity, conflicting preferences, the information overload found on the internet and the tourists' different ways of coping with the information, hindering the recommendation process from the users' profile construction to the final recommendation of a list of points of interest to visit. In this work, we show how we tackled the identified issues in a GRS prototype, Grouplanner, including the cold-start problem, by predicting the tourists' preferences based only on their personality and dividing the main group into subgroups of similar personality; by using a Multi-Agent Microservice; a novel dynamic clustering algorithm, d -means, adapted from the k -means algorithm, that does not need to know the number of clusters a priori; and association rules. Using a personality dataset of n =100k users, the proposed d -means algorithm was tested against two baselines (k -means and k -means++), showing better results in the clustering quality and scalability. We were also able to determine a large set of association rules to refine the recommendations, although further improvements are needed. To test the Grouplanner prototype, a simulation with real users (n =35) was conducted. The results showed the subgroups formed were very compact, revealing a very good clustering quality, with an average silhouette of s = 0.91. 11 of the 15 proposed tourist preferences were successfully predicted and used for the preliminary recommendation lists, being 92 % of the participants satisfied with the individual recommendations and 96 % with the group recommendations. [ABSTRACT FROM AUTHOR]

In the era of big data, information overload poses a challenge, complicating user decision-making. Recommender systems aim to assist in this process. In recent years, research on point-of-interest (POI) recommendations has been gaining momentum with some studies pointing to issues that need to be resolved. Previous studies often used heterogeneous graphs to learn across different entity types, overlooking same-type entity relationships. Certain studies solely extract raw node features from a single source, thus disregarding information diversity, whereas others employ inappropriate methods that fail to preserve the inherent characteristics of the relevant information in the design of raw inputs. The integration of multiple sources of information can introduce a certain amount of noise into the data; however, the approaches used in related research may not be effective in handling this situation. To address these issues, we propose a hybrid structural graph attention network (HS-GAT) for POI recommendation. In this approach, multisource data are first preprocessed and relevant raw features are initialized. Subsequently, heterogeneous graphs are built for user-POI-POI attributes and POI-user-user attributes. These heterogeneous graphs are aggregated using a dual-attention mechanism, to create embedding matrices for users and POIs, which are then used to construct user-user and POI-POI homogeneous graphs. These graph structures are then combined with user and POI embeddings obtained from heterogeneous graphs and fed into a graph attention network (GAT), which yields the final embedding representations for users and POIs. Finally, recommendations for POIs are made in the form of inner products. A comprehensive performance evaluation of HS-GAT on the Yelp, Boston, Chicago and London datasets demonstrated that the proposed approach outperforms other state-of-the-art methods. • Build heterogeneous and homogeneous graphs to represent users and POIs. • Using a two-stage GNN framework to learn node representations in both graph types. • Various information types use specific methods for designing raw features. • Integrate raw features using multi-channel convolution from diverse sources. • Experimental results on real datasets show this study surpasses other baselines. [ABSTRACT FROM AUTHOR]

The social network-based recommendation algorithm can not only effectively solve the information overload problem, but also fully exploit the personalized preferences of users. However, current studies fail to consider the distrust feature and propagation characteristic of social networks. In addition, methods for quantifying user preferences and social relationships are deficient and inconsistent with reality. Therefore, an extended trust and distrust network-based dual fuzzy recommendation model (ETD-DFR) is proposed. First, the trust and distrust network is constructed, and then the network is extended based on the social balance theory; secondly, the weighted PageRank algorithm is utilized to calculate the trust and distrust between users; Further, a dual fuzzy framework is proposed to represent the fuzzy features of user preferences as well as trust and distrust; In addition, Louvain's algorithm is employed to partition the network into subgroups to quickly find trusted users for recommendation. An empirical analysis is conducted based on the dataset of consumer review websites, and the experimental results show that the performance of the proposed ETD-DFR model is excellent. Finally, comparative and sensitivity analyses are constructed to further verify the proposed ETD-DFR model is efficient and robust. [ABSTRACT FROM AUTHOR]

This article presents a novel approach to enhancing personalized recommendations in online social networks by exploiting the concept of Social Capital. Recognizing the challenges posed by information overload on platforms like Twitter, the proposed method integrates user interactions and features on social media with the concept of Social Capital to generate more relevant recommendations. The model incorporates a multifaceted analysis of user data, including the user's reputation, influence, and engagement strength, alongside tweets' recency, diversity, and context scores, to calculate a comprehensive Social Capital Score for each recommendation. An extensive offline evaluation of the model demonstrates its effectiveness in improving the personalization and relevance of recommendations. The findings highlight the significant role of incorporating Social Capital in recommender systems, contributing to a more engaging and informative user experience on social media platforms. This study provides valuable insights into the potential benefits and limitations of this approach, advancing the field of recommender systems and enhancing our understanding of the role of social capital in shaping user behavior and preferences. [ABSTRACT FROM AUTHOR]

Fake news has evolved into a pervasive issue in the era of information overload, influencing public opinion and challenging the credibility of news sources. While various approaches have been proposed to combat fake news, most existing research focuses on high-resource languages, leaving low-resource languages vulnerable to misinformation. In this study, we propose a hybrid deep learning model architecture that integrates dilated temporal convolutional neural networks (DTCN), bidirectional long-short-term memory (BiLSTM), and a contextualized attention mechanism (CAM) to address the problem of detecting fake news in low-resourced Dravidian languages. DTCN is employed to capture temporal dependencies due to its sequential nature, BiLSTM is employed to seize long-range dependencies efficiently, and CAM is used to emphasize important information while downplaying irrelevant content. Additionally, we incorporate an adaptive-based cyclical learning rate with an early stopping mechanism to enhance model convergence. The results demonstrate that the proposed model surpasses the state-of-the-art and baseline models and achieves a higher average accuracy of 93.97% on the Dravidian_Fake dataset in four Dravidian languages. • We proposed a novel hybrid deep learning model for Dravidian fake news detection. • We proposed an adaptive learning rate method for fast convergence of the model. • We performed a comparative analysis of baseline models with the proposed model. [ABSTRACT FROM AUTHOR]

The Internet has revolutionised the way information is retrieved, and the increase in the number of users has resulted in a surge in the volume and heterogeneity of available data. Recommender systems have become popular tools to help users retrieve relevant information quickly. Food Recommender Systems (FRS), in particular, have proven useful in overcoming the overload of information present in the food domain. However, the recommendation of food is a complex domain with specific characteristics causing many challenges. Additionally, very few systematic literature reviews have been conducted in the domain on FRS. This paper presents a systematic literature review that summarises the current state-of-the-art in FRS. Our systematic review examines the different methods and algorithms used for recommendation, the data and how it is processed, and evaluation methods. It also presents the advantages and disadvantages of FRS. To achieve this, a total of 67 high-quality studies were selected from a pool of 2,738 studies using strict quality criteria. The review reveals that the domain of food recommendation is very diverse, and most FRS are built using content-based filtering and ML approaches to provide non-personalised recommendations. The review provides valuable information to the research field, helping researchers in the domain to select a strategy to develop FRS. This review can help improve the efficiency of development, thus closing the gap between the development of FRS and other recommender systems. • Most recommender systems use content-based filtering and predict recommendations with various machine learning algorithms. • Graph-based systems that employ various machine learning algorithms for recommendation are prevalent recently. • Several studies ignore personal attributes of users when producing recommendations. • Most systems use one source for their data, which is most often Allrecipes. • Offline evaluation with accuracy-based metrics is widely-used for measuring the performance of the FRS. [ABSTRACT FROM AUTHOR]

• CT-GNN: a graph neural network model for session-based recommendation incorporating temporal information. • LIRL module: learns users' local preference for better recommendations. • Time label preference: introducing temporal information into session-based recommendations for improved performance. • CT-GNN outperforms baseline models on two datasets, achieving high performance quickly. With the rapid growth of Internet data, recommendation systems have become the basic technology to alleviate information overload. The session-based recommendation (SBR) is a challenging task, and its purpose is to predict user behavior based on anonymous sessions. The existing SBR methods need to be improved in capturing the complex conversion relationship of items. In contrast, Graph Neural Network (GNN) can capture the complex conversion between items by modeling sessions as graph structure data. However, current methods just sort the clicked items within a session based on time, without utilizing the temporal information between sessions, leading to poor recommendation performance. To improve the accuracy of session recommendation (SR), we propose an SBR model based on GNN with Combined Temporal (CT-GNN) information. The proposed CT-GNN model is built based on the time of session occurrence, and it can learn the temporal association relationship between session items to enrich the connection between items. More importantly, based on the diversity problem faced by current SRs, which refers to the tendency of users to engage with popular items and resulting in limited exposure for other items, the CT-GNN model uses the Local Item Representation Learning (LIRL) module to learn users' local preferences. Through the LIRL module, the CT-GNN can capture users' interests, boost item exposure, and increase recommendation diversity. The experimental results show that the CT-GNN model is superior to the state-of-the-art SBR methods, with MRR scores 1.1 %∼3.4 % higher than the best-performing baseline; recall scores 0.9 %∼2.3 % higher than the best-performing baseline, and alleviates the diversity problem faced by SR, greatly improving the speed of model convergence, and the training time is 13.6 %∼20.7 % of the best-performing baseline. [ABSTRACT FROM AUTHOR]

In the age of information overload, modern recommendation systems provide an important role in helping people screen massive information. With the development of deep learning, personalized recommendations are further promoted to enhance user engagement and improve user satisfaction. Meanwhile, when the content provider has high-priority content to peddle, it often just exposes it to a conspicuous location without considering different users' preferences, which may even lead to negative impressions. To tackle this problem, in this paper, we propose a target-driven user preference transferring (TDUPTrans) module that will gradually orientate the user's preferences toward the target content. It is a brand new attempt that applies the user preference transfer process on personalized recommendations so that users don't just "see" the target item but also "like" it. To be specific, we propose a greedy-strategy-based intermediate content choosing algorithm, which can recommend different intermediate contents to users with different initial interests, and eventually lead users' interests to the target content. We give a clear interpretation of how our proposed process works through a news-based case study. Moreover, extensive experiments on real-world data sets demonstrate the interpretability of the guiding process, the abilities of our proposed TDUPTrans fusing with pre-existed recommendation algorithms, and the significantly improved target recommendation success rate with this TDUPTrans module plugged in. [ABSTRACT FROM AUTHOR]

Advanced organizational processes today require many process-relevant artifacts to be formally represented in process models. As these processes become more complex, process designers face the challenge of maintaining high comprehensibility while adequately visualizing process-relevant information. Modularization has been introduced to address the increasing complexity and information overload of process models, tackling this issue. While research acknowledges the importance of process model comprehensibility in order to benefit from the merits of such models, there still needs to be more clarity about the benefits of process model modularization. Recent research has focused on the human visual system and investigated visual behavior in the context of process model literacy. Continuing this branch of research, this paper analyses gaze behavior when comprehending modularized process models. A follow-up eye-tracking study was conducted to comprehend differently modularized BPMN 2.0 process models (i.e., vertical, horizontal, orthogonal). The results indicated that process model readers comprehend modularization approaches differently. Vertical modularization was perceived as the best comprehensible approach, whereas horizontal received the least acceptance. Considered eye movement performance measures confirm prior observation that attention is deployed more effectively in vertical modularization. Based upon the findings, the paper proposes a revised theoretical model from prior work that relates visual tasks to the execution of visual routines, explaining comprehension and information processing in modularized process models. The theoretical model elaborates how the visual system process and comprehends information obtained from modularized process models. Furthermore, the paper derived implications for modeling and practical applications, pointing out solid representational points and shortcomings of each modularization approach regarding how modularized process models are read and comprehended. [ABSTRACT FROM AUTHOR]

• A key review extraction method for e-commerce products is proposed. • A multi-objective optimization model for extracting reviews is constructed. • Product attribute weights and question weights are designed to extract reviews. • A binary multi-objective reptile search algorithm with RL and CL is constructed. • The superiority of the proposed method is shown through experiments. Online product reviews on e-commerce platforms help consumers evaluate and select products. However, the large number of accumulated product reviews leads to information overload and redundancy. It is difficult for consumers to quickly obtain needed information. Therefore, this study proposes a new model for extracting key reviews from the perspectives of supporting consumer purchase decisions and reducing review information overload. First, patterns of contextual information surrounding product attributes are extracted to model the product features, questions, and reviews in the same semantic space. Second, the new key review extraction problem is defined as a multi-objective optimization model. Five objective functions in terms of supporting consumer purchase decisions and reducing review information overload are defined in the model. Product attribute weights and product question weights are designed to construct the objective functions. Third, a binary multi-objective reptile search algorithm RLRSA-CL integrating reinforcement learning and cross-learning strategies is innovatively constructed to solve the optimization model and obtain key reviews. Finally, experiments on a real dataset from Jingdong Mall are performed. The results show that the proposed RLRSA-CL algorithm is valid and superior to other algorithms in key review extraction. [ABSTRACT FROM AUTHOR]

• A new biomedical text summarization based on clustering and frequent itemsets mining. • Biomedical texts are represented using concepts instead of terms. • This combination enhances the quality of the generated summaries. • The clustering has a crucial impact on the discovered frequent itemsets. • Contribute all clusters in the sentence selection step yields to better performances. Text summarization has become an important research area, especially in the biomedical domain, where information overload is a major problem. In this paper, we propose a novel biomedical text summarization system that combines two popular data mining techniques: clustering and frequent itemset mining. Biomedical paper is expressed as a set of biomedical concepts using the UMLS metathesaurus. The K-means algorithm is used to cluster similar sentences. Then, the Apriori algorithm is applied to discover the frequent itemsets among the clustered sentences. Finally, the salient sentences from each cluster are selected to build the summary using the discovered frequent itemsets. For the evaluation step, we selected randomly 100 biomedical papers from the BioMed Central database full-text , and we evaluated the performances of our system by comparing the resulting summaries with the abstracts of these papers using the ROUGE metrics in term of recall, precision, and F-measure. We also compared the obtained summaries with those achieved by five well-known summarizers: TextRank, TextTeaser, SweSum, ItemSet Based Summarizer, Microsoft AutoSummarize , and two baselines: summarization using only the frequent itemsets mining (FRQ-CL), and summarization using only the clustering (CL-FRQ). The results demonstrate that this combination can successfully enhance the summarization performances, and the proposed system outperforms other tested summarizers. [ABSTRACT FROM AUTHOR]

Highlights • A fast recommendation approach is proposed. • The approach is based on self-adaptation and multi-thresholding. • An automatic convergence procedure is proposed to improve the performance of our approach. • The MAE is improved between 1.02% and 12.93%, and the computational time between 25.38% and 54.83%. Abstract Finding relevant and interesting items on e-commerce websites is a major challenge in the information era. Recommender systems are widely used to help users deal with the information overload problem by giving them personalized recommendations. Furthermore, they help businesses make more profits. Collaborative filtering represents one of the most successful recommendation approaches. In general, several existing recommendation methods have demonstrated good performance in terms of prediction quality. However, they may require prohibitive computational times, and they often confront the sparsity problem, which negatively affects the efficiency of the system. In this paper, we propose a fast recommendation algorithm based on self-adaptation and multi-thresholding. It is designed to deal with the mentioned drawbacks, and improve the prediction quality. In addition, it is able to converge automatically. Extensive experiments on two real-world data sets demonstrate that our proposed method can achieve significantly better performance than other state-of-the-art methods. Particularly, it improves the MAE between 1.02% and 12.93%, and the computational time between 25.38% and 54.83%. [ABSTRACT FROM AUTHOR]

Recommender systems were first introduced to solve information overload problems in enterprises. Over the last few decades, recommender systems have found applications in several major websites related to e-commerce, music and video streaming, travel and movie sites, social media, and mobile app stores. Several methods have been proposed over the years to build recommender systems. However, very little work has been done in recommender system evaluation metrics. The most common approach to measuring recommender system's performance in offline settings is to employ micro or macro averaged versions of standard machine-learning measures. Profit or other business-oriented metrics have been proposed for other predictive analytics problems, such as churn prediction. However, no such metrics have emerged for the recommender system context. In this work, we propose a novel evaluation metric that incorporates information from the online-platform userbase's behavior. This metric's rationale is that the recommender system ought to improve customers' repeated use of an online platform beyond the baseline level (i.e. in the absence of a recommender system). An empirical application of this novel metric is also presented in a real-world mobile app store, which integrates the dynamics of large-scale big data environments, which are common deployment scenarios for these types of recommender systems. The resulting profit metric is shown to correlate with the existing metrics while also being capable of integrating cost information, thereby providing an additional business benefit context, which allows us to differentiate between two similarly performing models. • We introduce a business view on recommender system evaluation metrics (NBD-Dirichlet). • We propose a novel metric that includes information from the online platform user's behavior. • We presented an application of this metric using a real-world mobile app store large-scale dataset. • The results presented have the advantage of including infrastructure cost information.. [ABSTRACT FROM AUTHOR]

Recommender systems have been developed to overcome the information overload problem by retrieving the most relevant resources. Constructing an appropriate model to estimate the user interests is the major task of recommender systems. The profile matching and latent factors are two main approaches for user modeling. Although a notion of timestamps has already been applied to address the temporary nature of recommender systems, the evolutionary behavior of such systems is less studied. In this paper, we introduce the concept of trend to capture the interests of user in selecting items among different group of similar items. The trend based user model is constructed by incorporating user profile into a new extension of Distance Dependent Chines Restaurant Process (dd-CRP). dd-CRP which is a Bayesian Nonparametric model, provides a framework for constructing an evolutionary user model that captures the dynamics of user interests. We evaluate the proposed method using a real-world data-set that contains news tweets of three news agencies (New York Times, BBC and Associated Press). The experimental results and comparisons show the superior recommendation accuracy of the proposed approach, and its ability to effectively evolve over time. [ABSTRACT FROM AUTHOR]

Recommender systems have emerged as a key tool to overcome the negative impact of information overload problem, as well as, help the users seek the relevant information based on their past preferences. Collaborative filtering represents a widely used approach to build recommendation systems. In essence, many methods have been developed to provide high quality results, neverthless, they may incur prohibitive computational costs. In this paper, a novel method called FRAIPA is proposed, which is designed to tackle the sparsity, dynamic data problems, moreover, it improves the prediction accuracy and computation time. Experimental results on two real-world data sets reveal the effectiveness of the proposed method over existing methods. [ABSTRACT FROM AUTHOR]

Recommender systems have become a pervasive and effectual means to cope with the boosted complexity of decision-making process resulting from information overload. The key factor confronting recommendation tasks is to accurately characterize user preferences which are not only intrinsically dynamic and evolving, but also associated with multiple features and contexts. Sequential models especially Recurrent Neural Networks (RNNs) have become a powerful tool to address the issue. However, most of current RNN methods select the sequence of items that the user interacted with in the past to represent the current user state without considering contextual information associated with users and items simultaneously. Besides, the existing methods are hard to access and store the importance information of user representation. In this paper, we put forward two types of Gated Recurrent Units (GRU) models, i.e., multi-feature linear fusion GRU and multi-feature user integration GRU, to address the above drawbacks and thus help recommend more personalized next items to users. The proposed models employ multi-feature fusion technology including feature concatenation and one-dimensional Convolutional Neural Network (1D CNN) to generate input of GRU for learning more feature interaction information. Experiments on two real-world datasets demonstrate the performance gains of the proposed model compared with state-of-the-art. Further ablation studies verify the impact of different features for two models. • A linear GRU with multi-feature fusion based on contextual information is proposed. • A new multi-feature user integration GRU to filter user information is proposed. • The TextCNN is used to handle the diversity of text data in the datasets. • Ablation studies verify the impact of different features. • Our methods outperform competitive methods in terms of MRR, Recall and F1. [ABSTRACT FROM AUTHOR]

Due to the rapid growth of the information overload issue, recommender systems have become necessary and are implemented in numerous facets of human life, including the tourism industry. Today, technological advancements have significantly altered our travels, and these advancements promise even more interactive and exciting experiences in the future. Nowadays, planning and arranging a customized trip well in advance is ideal, as the process can be challenging and time-consuming. This paper proposes a hybrid approach that leverages multidimensional data to enhance personalized trip recommendations while addressing several shortcomings of existing recommender systems, such as their inability to account for dynamic user preferences and diverse contexts. To this end, the proposed method, considering the data sparsity issue, employs a clustering algorithm to reduce the time complexity of discovering Points of Interest (POIs). This approach utilizes user demographic information to address the cold start issue while improving the collaborative filtering (CF) paradigm through an asymmetric schema. Furthermore, this study exploits the context vector model and the Term-Frequency-Inverse-Document-Frequency (TF-IDF) algorithm to present a novel method for quantifying context similarity. Moreover, the method retrieves and ranks a list of candidate routes optimized by applying personalized POIs to sequential travel patterns. Finally, the experimental results demonstrate the superiority of the approach in terms of Precision, Recall, RMSE, MAP, F-Score, and nDCG compared to previous works based on Flickr and STS datasets. [ABSTRACT FROM AUTHOR]

• We propose a novel research-related recommendation system using BERT and LDA. • We propose an explainable keywords extractor using BERT's self-attention mechanism. • Our method provides an effective research-related information retrieval process. With the launch of academic search engines such as Google Scholar, Microsoft Academic, and Scopus, researchers are using them to access scholarly materials on a large scale without making any payment. In this optimal research environment, academic libraries or community databases have naturally experienced a rapid increase in volume; however, the excessive quantitative growth of information, or information overload, acts as a double-edged sword that prevents researchers from finding relevant prior studies or researchers with similar interests. Existing keyword and rule-based search systems carry the risk of recommending research literature with a high citation frequency rather than recommending contextually similar documents. Therefore, this study proposes a semi-supervised semantic-based research literature and researcher recommendation system using LDA and BERT. Since a semi-supervised method is used, research literature can be embedded based on contextual and classification information, and the global topic information of the research literature can also be captured. In addition, a research literature information extractor system has been implemented, which comprises classification network of our model and an explainable keywords extractor system implemented using BERT's self-attention structure. Based on the experimental results, it can be confirmed that the proposed study model shows better performance than other baselines. This model can help users to find information and contextual matching data quickly and accurately from the database. In addition, when implementing an academic library and community database, such a powerful research-related recommendation system and research literature information extractor are expected to have an enormous effect. [ABSTRACT FROM AUTHOR]

Abstract: Users face the information overload problem when downloading applications in markets. This is mainly due to (i) the increasing unmanageable number of applications and (ii) the lack of an accurate and fine-grained categorization of the applications in the markets. To address this issue, we present an integrated solution which recommends to the users applications by considering a big amount of information: that is, according to their previously consumed applications, use pattern, tags used to annotate resources and history of ratings. We focus this paper on the service for monitoring users’ interaction. [Copyright &y& Elsevier]

Abstract: Recommender systems are popular tools dealing with the information overload problem in e-commerce web sites. The more they know about the users, the better recommendations they can provide. However, sometimes, in real situations, it is necessary to make guesses about the value of missing but useful data in order to generate a recommendation immediately, rather than waiting the data becomes available. This paper presents an assumption-based multiagent recommender system capable of making these types of assumptions about the preferences of the users. The approach was validate in the tourism domain (recommendation of travel packages). Experiments were conducted to illustrate the impact of various assumption making strategies on the quality of the recommendations as well as the impact of trust assignment. [Copyright &y& Elsevier]

Abstract: The improvement of information technology makes storage no longer a problem. In addition, the birth of the Internet makes information transfer faster than ever. It brings us convenient life. However, more and more information result in a new problem, which is information overload. Today, many more people are traveling abroad since they no longer have to work on weekends. Traveling abroad has become a kind of trend. There are more than a hundred countries in the world worth to travel, and there is so much information available that it makes a traveler’s decision extremely difficult to make. In our research, we try to implement the most common three kinds of recommender system techniques in order to recommend to customers which countries are the best traveling locations for them. Thus, we can save travelers a lot of time when deciding where to go. From our experiment and evaluation, we find that a hybrid recommender system is a better technique in recommendation according to our abroad database, and it conquers the shortcomings of content-based filtering and collaborative filtering approaches. [Copyright &y& Elsevier]

Abstract: Most of current knowledge based systems manage impressive amounts of information (especially distributed fuzzy information). In addition to widely pointed-out integration and maintenance difficulties, other common problem is overwhelming of users with much more information than the strictly necessary for fulfilling a task. This issue has been pointed out with the name of “information overload”. Use of context knowledge has been envisioned as an appropriate solution to deal with this information overload matter. In this paper, we present a distributed fuzzy context-domain relevance (DFCDR) model for representation in fuzzy ontologies relevance relations between fuzzy context ontology and distributed fuzzy domain ontologies. In fact, the DFCDR model is a distributed fuzzy extension of the context-domain relevance (CDR) model. [Copyright &y& Elsevier]

Abstract: Different efforts have been done to address the problem of information overload on the Internet. Recommender systems aim at directing users through this information space, toward the resources that best meet their needs and interests by extracting knowledge from the previous users’ interactions. In this paper, we propose three algorithms to solve the web page recommendation problem. In our first algorithm, we use distributed learning automata to learn the behavior of previous users’ and recommend pages to the current user based on learned patterns. By introducing a novel weighted association rule mining algorithm, we present our second algorithm for recommendation purpose. Also, a novel method is proposed to pure the current session window. One of the challenging problems in recommendation systems is dealing with unvisited or newly added pages. By considering this problem and improving the efficiency of first two algorithms we present a hybrid algorithm based on distributed learning automata and proposed weighted association rule mining algorithm. In the hybrid algorithm we employ the HITS algorithm to extend the recommendation set. Our experiments on real data set show that the hybrid algorithm performs better than the other algorithms we compared to and, at the same time, it is less complex than other proposed algorithms with respect to memory usage and computational cost too. [Copyright &y& Elsevier]

As customers become more skilled in the use of internet, many companies have gradually established their websites with more and more enormous information to get future competition in electronic commerce (EC). However, the miscellaneous information often brings the users at a loss. Web personalization provides a solution to improvement of information overloading on websites. The objective of web personalization is to give users a website they want or need, and thus knowing the needs of users is an important task for content recommendation in web personalization. In this article, we propose a hybrid approach for this task. The proposed approach trains the artificial neural networks to group users into different clusters, and applies the well-established Kano’s method to extracting the implicit needs from users in different clusters. Finally, a real case of tour and travel websites applying the approach is presented to demonstrate the improvement of information overloading. [Copyright &y& Elsevier]

Precise Short term load forecasting (STLF) plays a significant role in the management of power system of countries and regions on the grounds of insufficient electric energy for increased need. This paper presents an approach of back propagation neural network with rough set (RSBP) for complicated STLF with dynamic and non-linear factors to develop the accuracy of predictions. Through attribute reduction based on variable precision with rough set, the influence of noise data and weak interdependency data to BP is avoided so the time taken for training is decreased. Using load time series from a practical power system, we tested the performance of RSBP by comparing its predictions with that of BP network. [Copyright &y& Elsevier]

To help the growing qualitative and quantitative demands for information from the WWW, efficient automatic Web page classifiers are urgently needed. However, a classifier applied to the WWW faces a huge-scale dimensionality problem since it must handle millions of Web pages, tens of thousands of features, and hundreds of categories. When it comes to practical implementation, reducing the dimensionality is a critically important challenge. In this paper, we propose a fuzzy ranking analysis paradigm together with a novel relevance measure, discriminating power measure (DPM), to effectively reduce the input dimensionality from tens of thousands to a few hundred with zero rejection rate and small decrease in accuracy. The two-level promotion method based on fuzzy ranking analysis is proposed to improve the behavior of each relevance measure and combine those measures to produce a better evaluation of features. Additionally, the DPM measure has low computation cost and emphasizes on both positive and negative discriminating features. Also, it emphasizes classification in parallel order, rather than classification in serial order. In our experimental results, the fuzzy ranking analysis is useful for validating the uncertain behavior of each relevance measure. Moreover, the DPM reduces input dimensionality from 10,427 to 200 with zero rejection rate and with less than 5% decline (from 84.5% to 80.4%) in the test accuracy. Furthermore, to consider the impacts on classification accuracy for the proposed DPM, the experimental results of China Time and Reuter-21578 datasets have demonstrated that the DPM provides major benefit to promote document classification accuracy rate. The results also show that the DPM indeed can reduce both redundancy and noise features to set up a better classifier. [Copyright &y& Elsevier]

Abstract: Bioinformatics is a rapidly developing field of quantitative biomedical research with the potential to transform the three principle domains within healthcare operations of research, clinical practice and administration. Presently, a significant part of bioinformatics research concentrates on genomics and proteomics. Despite growing clinical enthusiasm for the discoveries and creation of new information and knowledge that such research brings, current results are still outside the daily clinical reality. As a measure to close the gap and integrate bioinformatics into the rest of the healthcare domains, the authors propose a networkcentric approach. Based on operations within the unified space created by the overlap of three domains of knowledge, networkcentric healthcare operations support free information flow among all constituents (actors) within the healthcare space, rapid generation and exchange of pertinent knowledge, as well as enhanced awareness of the significance and practical implementation of new discoveries within specialized fields of biomedicine (e.g. bioinformatics). Healthcare networkcentricity will also facilitate conversion of information into readily accessible knowledge, accelerate translation of that knowledge into clinical practice, and reduce the stress of information overload. Changes in the use of Information/Computer/Communication Technologies (IC2T) and enhanced efficiency of Decision/Executive Support Systems (DSS/ESS) are pivotal to the success of the networkcentric approach as discussed in the paper. [Copyright &y& Elsevier]

• Accurate and Novel Session-based News Recommendations. • Development of a stream-based version of SimRank algorithm. • Providing News Suggestions over Time-evolving Graphs. Recommender systems are among the most widespread applications of artificial intelligence techniques. For instance, news recommender systems serve users in managing the overload of information they come across when accessing news portals. Obviously, in the news domain time-awareness of recommendation approaches are crucial. However, most of these approaches missed to consider user sessions, which group the items that a user interacted with. In this paper, we study the problem of session-based recommendations by running SimRank on time-evolving heterogeneous graphs. In particular, we construct a dynamic heterogeneous multi-partite graph and adjust SimRank to run on it by using different (i) sliding time window sizes, (ii) sub-graphs used for model learning and (iii) sequential article weighting strategies. We evaluate our algorithms on two real-life datasets, and we show that our method outperforms other state-of-the-art methods in terms of accuracy and diversity. The significance and impact of this work is important because it introduces to the research community of expert and intelligent systems, for the first time, a stream-based version of SimRank algorithm, which is able to run over time-evolving graphs. [ABSTRACT FROM AUTHOR]

The amount of available multimedia data in different formats and from different sources increases everyday. From an information retrieval point of view, this high volume and heterogeneity of data involves several issues to be addressed related to information overload and lacks of well structured information. Even if modern information retrieval systems offer to the user manifold search options, it is still hard to find systems with optimal performances in the document seeking process starting from a given topic. In recent years, several frameworks have been proposed and developed to support this task based on different models and techniques. In this paper we propose a semantic approach to document classification using both textual and visual topic detection techniques based on deep neural networks and multimedia knowledge graph. A semantic multimedia knowledge base has been exploited and several experimental results show the effectiveness of our proposed approach. • A combination of textual and visual information is suitable for topic detection. • Semantic analysis and deep neural networks are used in our approach. • Knowledge is represented by a Multimedia Knowledge Graph. • A web document collection is a real scenario to test our strategy. [ABSTRACT FROM AUTHOR]

• Analysis the parameters adjustment range of the traditional method. • A dual parameters method was proposed to expand the range. • The best parameters selection is obtained by particle swarm optimization. • Our method has better balance between accuracy and diversity. With the unprecedented development in the internet technology, the information overload issues have become more and more complex, resulting in users being unable to obtain the target information accurately and effectively in selecting the required information from a large pool of surfed data. In view of this a recommendation system can be used to predict the user's selection probability for different potential objects as an important tool, which can help to solve the information overload issues. So far, many personalized recommendation algorithms based on bipartite graphS have been proposed, most of which are based on the similarity degree among users or items, such as collaborative filtering (CF), mass diffusion (MD) and heat conduction (HC). Among many recommendation algorithms, the performances of algorithms are varied. MD algorithm has high recommendation accuracy but poor diversity, while HC algorithm has good diversity but low accuracy. In order to solve the dilemma in accuracy and diversity, some hybrid recommendation algorithm have been proposed. This paper has mainly focused on the hybrid recommendation algorithm HHM, and pointed out its shortcomings. Based on the reconsideration of the effect of item popularity in the recommendation process, an improved hybrid recommendation algorithm using dual parameter called IHM was proposed. The particle swarm optimization (PSO) algorithm was applied to the parameter optimal of the hybrid recommendation algorithm to obtain the parameters of the algorithm. Experiments on 3 real datasets indicated that the IHM algorithm is better than HHM algorithms in terms of the recommendation accuracy, diversity and novelty. Meanwhile, the IHM algorithm can also improve the recommendation for items with lower popularity and solve the cold start problem. [ABSTRACT FROM AUTHOR]

• Storage and representation of fuzzy spatiotemporal data objects on XML. • Query relaxation combining proximity and fuzziness of fuzzy spatiotemporal data. • Result organization considering data distribution and user context preferences. • A framework of query relaxation, result categorization and navigation tree construction. With the rapid development of spatiotemporal information and its applications, querying spatiotemporal data has received considerable attention. Meanwhile, the imprecision and uncertainty of information cannot be ignored in many practical applications. Querying fuzzy spatiotemporal data have become one of the most important topics in academia and industry. Although there have been some achievements in querying aspect, the study about fuzzy spatiotemporal query relaxation is still few. In fact, query relaxation is necessary when the amount of query results is small or even empty, especially in the process of querying fuzzy spatiotemporal data. In this paper, we propose an adaptive query relaxation and result categorization approach for fuzzy spatiotemporal data based on XML, which is compatible with spatiotemporal features when spatiotemporal related queries are performed. The approach does not depend on any specific domain or user, it can adaptively relax the initial query requirements, and classify the results by user context preferences and data distribution after query relaxation. In order to locate the spatiotemporal data quickly and show the corresponding fuzziness apparently, we adopted XML to construct the fuzzy spatiotemporal model for query relaxation because XML data can be represented as a tree model which can flexibly arrange the spatiotemporal nodes at the specified position and mark the fuzziness of nodes on the path. In addition, after query relaxation, we present the results categorization algorithm to address the problem of information overload, and then return a navigation tree to the user. Finally, we launch a comprehensive set of experiments to demonstrate the effectiveness and efficiency of our proposed approach. Results of experiments demonstrate that our adaptive query relaxation and result categorization approach based on XML has higher recall and precision in spatiotemporal related query, and can capture the user's needs and preferences effectively as well. [ABSTRACT FROM AUTHOR]

With the development of mobile Internet, microblog has become one of the most popular social platforms. The enormous user-generated microblogs have caused the problem of information overload, which makes users difficult to find the microblogs they actually need. Hence, how to provide users with accurate microblogs has become a hot and urgent issue. In this paper, we propose an approach of hybrid microblog recommendation, which is developed on a framework of deep neural network with a group of heterogeneous features as its input. Specifically, two new recommendation strategies are first constructed in terms of the extended user-interest tags and user interest topics, respectively. These two strategies additionally with the collaborative filtering are employed together to obtain the candidate microblogs for final recommendation. Then, we propose the heterogeneous features related to personal interests of users, interest in authors and microblog quality to describe the candidate microblogs. Finally, a deep neural network with multiple hidden layers is designed to predict and rank the microblogs. Extensive experiments conducted on the datasets of Sina Weibo and Twitter indicate that our proposed approach significantly outperforms the state-of-the-art methods. The code and the two datasets of this paper are publicly available at GitHub. • Hybrid microblog recommendation using deep neural network framework is proposed. • The two strategies are proposed to extend user interest tags and extract interest topics for obtaining candidate recommended microblogs. • A group of heterogeneous features are proposed to represent microblogs. • A deep neural network with multiple hidden layers is designed to rank microblogs. • Extensive experiment results on two datasets of Twitter and Sina Weibo outperform the state-of-art methods. [ABSTRACT FROM AUTHOR]

• Revealing research hotspot and corresponding characteristics by bibliometric methods. • Discovering underlying laws behind data by constructing scientific knowledge graph. • Summarizing potential research hotspots by keyword co-occurrence cluster graph. • Summarizing cutting-edge trends by keyword co-occurrence cluster graph. • Discussing the main open problems in depth and proposing corresponding solutions. With the advent of the era of big data, the recommendation system has become an effective solution to the problem of information overload. This paper takes the literature data related to the recommendation system theme from 2009 to 2018 and included in the core collection of Web of Science database as the research object, and utilizes bibliometric methods to analyze the theme of recommendation system. To this end, firstly, classify statistics and feature analysis of valid literature data. Secondly, use VOSviewer software to construct various different scientific knowledge graph to discover valuable knowledge. Thirdly, according to keyword co-concurrence graph conclude five main hotspots of current research about recommendation system and discover five main directions that have potential value in research field of recommendation system. Finally, deeply explore five main key issues and propose corresponding solutions. [ABSTRACT FROM AUTHOR]

• Social influence could capture the intricate relations between users. • Item clustering could reflect user's preference. • Recommendation within each item cluster achieves high performance. • The incremental update strategy guarantees flexible online scalability. • Computational cost decreases with online incremental update. With the ever-increasing scale and complexity of social network and online business, Recommender Systems (RS) have played crucial roles in information processing and filtering in various online applications, although suffering from such as data sparsity and low accuracy problems. Meanwhile, recent researches try to enhance the performance of RS through such social network and clustering algorithms, however, they may fail to achieve further improvement in large-scale online recommendation due to the serious information overload. In this article, a novel social recommendation approach with online update referred to as LsRec is proposed, which generally contains offline computation and online incremental update. More precisely, LsRec not only takes account of user's social relationship, but also clusters items according to the similarity degree, furthermore, LsRec performs recommendation in each generated cluster respectively. In practice, LsRec could be capable of exploiting user-level social influence, and capturing the intricate relationship between items. In addition, theoretical proof could provide convergence guarantee for the model. Specifically, with the appealing merit of flexible online update scenario, LsRec could yield high performance in large-scale online recommendation with low computational complexity. Extensive experimental analysis over four real world datasets demonstrate the effectiveness and efficiency of LsRec, which indicates that LsRec could significantly outperform state-of-the-art recommender approaches, especially in large-scale online recommendation. [ABSTRACT FROM AUTHOR]

• A practical system for human rights monitoring combining NLP and crowdsourcing. • Mining social media offers signals for human rights abuses in addition to reports. • Deep learning outperforms traditional machine learning in our classification task. • The Ceasefire Iraq platform has been continuously applied for several years. Effective information management has long been a problem in organisations that are not of a scale that they can afford their own department dedicated to this task. Growing information overload has made this problem even more pronounced. On the other hand we have recently witnessed the emergence of intelligent tools, packages and resources that made it possible to rapidly transfer knowledge from the academic community to industry, government and other potential beneficiaries. Here we demonstrate how adopting state-of-the-art natural language processing (NLP) and crowdsourcing methods has resulted in measurable benefits for a human rights organisation by transforming their information and knowledge management using a novel approach that supports human rights monitoring in conflict zones. More specifically, we report on mining and classifying Arabic Twitter in order to identify potential human rights abuse incidents in a continuous stream of social media data within a specified geographical region. Results show deep learning approaches such as LSTM allow us to push the precision close to 85% for this task with an F1-score of 75%. Apart from the scientific insights we also demonstrate the viability of the framework which has been deployed as the Ceasefire Iraq portal for more than three years which has already collected thousands of witness reports from within Iraq. This work is a case study of how progress in artificial intelligence has disrupted even the operation of relatively small-scale organisations. [ABSTRACT FROM AUTHOR]

• This work introduces a new short-text conversational dataset in Spanish. • We propose a novel task for recommending users to join conversations. • We design and evaluate a new state-of-the-art neural recommender model. The massive amounts of data on social media networks can be overwhelming for users; for this reason, recommending relevant content becomes an essential task to avoid information overload. In this paper, we propose a new task for recommending users that might be interested in join conversations on specific domains. To that end, we introduce a new corpus that contains conversations threads from popular users on Twitter on domains such as politics, sports, or humanitarian activism. Modeling short-text conversations on microblogs can be difficult because user-generated data is unstructured and noisy. Previous works focused on recommending content to users based on latent factors models and collaborative filtering methods. We propose a state-of-the-art recommendation model based on a sequence-to-sequence neural architecture that encodes the text of users' profiles and the conversations' context using several variants of recurrent neural networks. The experimental results show that our method provides as much as 20% higher recall compared to baseline methods. Moreover, we use an end-to-end learning framework that allows downstream applications to use recommender systems (RSs) that generalize better to new content by using pre-trained embeddings, thus being useful across domains or events. [ABSTRACT FROM AUTHOR]

• Fight information overload by semantics-driven filtering and knowledge generation. • A new way of learning History based on natural language processing and Linked Data. • Customized collecting of texts to train neural networks and check user knowledge. • Find and assess relevant interconnections among semantic entities by word vectors. Organizations that preserve and promote heritage must meet the expectatives of sophisticated visitors who, far from wanting simply to be informed, desire to explore engaging and innovative technology-driven experiences which consider their particular interests and encourage them to discover more. We describe an approach based on quiz games that can be exploited in the deployment of such challenging experiences. The game consists of raising multiple-choice questions about a particular theme which is introduced by a Humanities expert through a brief narrative. Given the input text, a question and its right answer, our strategy provides the expert with a set of wrong alternatives (called distractors). These options are chosen from a (semi)automatically-built tailor-made corpus of documents by considering each player's level of knowledge on the game theme and exploiting Linked Open Data initiatives and natural language processing. On the one hand, automatic selection of distractors assists the Humanities expert to create games about very diverse topics without needing to be a specialist in all of them. On the other one, distractors are related to the right answer of each question in an appealing and meaningful way, which contributes to arouse the visitors' curiosity and their possible interest in exploring similar experiences in future visits. The work has been experimentally validated, achieving better results than a previous distractor identification strategy. [ABSTRACT FROM AUTHOR]

• Paper proposes using semantics as feature for text summarization. • Results prove that using semantics as feature improves system performance. • Evaluation on standard dataset shows competitive superiority. Automatic text summarization essentially condenses a long document into a shorter format while preserving its information content and overall meaning. It is a potential solution to the information overload. Several automatic summarizers exist in the literature capable of producing high-quality summaries, but they do not focus on preserving the underlying meaning and semantics of the text. In this paper, we capture and preserve the semantics of text as the fundamental feature for summarizing a document. We propose an automatic summarizer using the distributional semantic model to capture semantics for producing high-quality summaries. We evaluated our summarizer using ROUGE on DUC-2007 dataset and compare our results with other four different state-of-the-art summarizers. Our system outperforms the other reference summarizers leading us to the conclusion that usage of semantic as a feature for text summarization provides improved results and helps to further reduce redundancies from the input source. [ABSTRACT FROM AUTHOR]

• A novel knowledge map for question and answer archives is constructed. • A novel growing hierarchical 2-dimensional self-organizing map is proposed. • Extracting typical Q&A pairs is proposed to help users quickly grasp main contents. • A merging mechanism is proposed during vertical expansion to avoid parsing. • Experimental results show that the knowledge map is feasible and performs well. The rapid increase in the number of community-based question-and-answer services has built up large archives of questions and answers. These archives deliver a plethora of valuable knowledge to users who primarily browse to locate information. Question–answer pairs, which not only include knowledge content but also indicate knowledge needs, are a new form of information presentation. To facilitate browsing question and answer archives and alleviate information overload during the browsing process, this paper constructs a knowledge map for question and answer archives by exploiting question–answer pair characteristics to determine a more precise location of question–answer pairs. First, the questions and answers are modeled and then the knowledge map structure is completed. Questions and answers are the two main dimensions of this knowledge map, and their intersection comprises a cluster of corresponding question–answer pairs. The knowledge map can be widely and deeply extended to provide a comprehensive representation of the question and answer archive. When labeling the knowledge map to aid in interpretation and understanding, the key words are identified, and typical question–answer pair extraction methods are proposed. Finally, we conduct extensive experiments on a real dataset, and the results show that the proposed approach is feasible and performs well. [ABSTRACT FROM AUTHOR]