Search

In the development of applications for mobile multimedia computing, it has been a core problem to improve the efficiency of applications development. The use of Web APIs has played an important role in applications development and mashups creation, but it takes a lot of time to find a suitable Web API. In this paper, we propose to employ APIs recommendation to solve this problem. We use the matrix factorization (MF) model to predict the relationship between user and Web APIs. The relationship between user and Web APIs is represented by latent features. But different from MF model, we find that the relationship between Web APIs have an impact on prediction performance. This paper uses the embedding technique and mines the latent relationships between APIs and we develop a novel prediction model that leverages the latent relationships between APIs, which is named as LRM (LRM is short for latent relationships mining). The latent relations are input into the LRM model to produce the relation prediction between APIs and developers. The experimental results show that our method achieves high performance and behaves the best compared to existing baselines. [ABSTRACT FROM AUTHOR]

Due to the advent of IoT and the increasing interest of billions of Internet users towards video contents, a huge multimedia flows has been generated, and as a consequence, a massive load is applied on the underlying core network. This change can affect the network stability and lead to potential performance degradation (such as congestion and delays). This is because multimedia flows are hungry in bandwidth, and also the classical routing protocols currently used in IoT core network (like OSPF) are not adapted yet to support the heavy and the large scale of multimedia traffics with a good quality of service (QoS). In this paper, we introduce the multicommodity-based routing by proposing two contributions, Maximum Concurrent Flow Protocol (MCFPr) and Cache-based Maximum Concurrent Flow (C-MCF). They are conceived based on the Maximum Concurrent Flow approach in order to optimize the routing of multimedia data in the backbone of IoT networks. Both contributions are studied and compared with the state-of-art approaches under different scenarios showing good results, especially in the number of data packets sent (improvement by 50%), and in the transmission time (50% faster compared to the majority), which makes them promising solutions for a rapid and efficient routing in IoT core networks. [ABSTRACT FROM AUTHOR]

Scholarly big data refer to the rapidly growing scholarly source of information, including a large number of authors, papers, and massive scale scholarly networks. Extracting the profile attributes for Web users is an important step in Web user analysis. For Web scholarly users, profile attributes extraction should integrate multi-source and heterogeneous information resources. However, the traditional extraction models have two main drawbacks: (1) The traditional models require manual feature selection based on specific domain knowledge; (2) The traditional models cannot adapt to the diversities of Scholarly Web pages and cannot discover the relationships between different target entities which are far apart in different domains. To address these issues, we propose a profile attributes extraction model, PAE-NN, based on a Bi-LSTM-CRF neural network. This model can automatically extract the characteristics and contextual representations of each extracting entity through a Recurrent Neural Network with end-to-end training. It takes advantage of the long-memory sequence characteristics of LSTM network to effectively discover the long-term dependencies on extracting entities. Our experimental results on published datasets from the SMPCUP2017 Open Academic Competition and Aminer demonstrate that the proposed PAE-NN model outperforms existing models in terms of extraction precision, recall, and F1-score with large-scale training data. [ABSTRACT FROM AUTHOR]

Social media popularity prediction refers to using multi-modal content to predict the popularity of a post offered by an internet user. It is an effective way to explore advanced forecasting trends and make more popularity-sensitive strategic decisions for the future. Existing methods attempt to explore various multi-model features to solve this task, which only focus on local information, lacking global understanding for the post's content. In this paper, we propose social media popularity prediction with caption (SMPC), a novel architecture that integrates the caption as the global representation into the existing multi-model-feature-based popularity prediction method. To make good use of the generated captions, we process them in word-level, sentence-level and length-level ways, obtaining three kinds of caption features. To incorporate caption features, we exploit seven variants of the architecture by concatenating features in all the possible manners, for the feature fusion and training different combinations for the CatBoost regression. Extensive experiments are conducted on Social Media Prediction Dataset (SMPD) and show that the proposed approaches can achieve competing results against state-of-the-art models. [ABSTRACT FROM AUTHOR]

Cloud storage systems have been turned into the primary services of Internet users nowadays. While the application of such systems is exponentially increasing, deduplication algorithms help face scalability issues. Although source-side deduplication optimizes both storage and bandwidth, the main concern that deduplication algorithms suffer from is still data confidentiality. Message-locked encryption (MLE) is a well-known key management framework for secure deduplication to provide confidentiality. This framework is the basis of almost all the proposed secure deduplication solutions. Even though there are lots of literature works trying to provide secure deduplication algorithms, to the best of our knowledge, none of them provide an effective anonymity service for data owners. In this paper, we propose an N-anonymity algorithm to provide an effective anonymity service, capable of prohibiting even the cloud storage provider from knowing which users are storing the same data. The algorithm is analytically studied, and the results are validated by exhaustive implementations using real data. Furthermore, we propose an ID-based key management algorithm as the cornerstone of the secure cloud storage system. The proposed algorithm, which could be considered as an asymmetric extension of MLE, is easy to implement and compatible with the existed cloud architectures as well as the proposed anonymity-based deduplication system. [ABSTRACT FROM AUTHOR]

In recent years, Online Social Networks like Facebook, Twitter have become an integral part of our daily life to share a variety of information with others. These information contents are accessible to a large group of Internet users causing a serious privacy risk. In this paper, we propose a simple and flexible access control model to regulate access to users' resources on Twitter. Our proposed model employs an object matrix and access capability list to enable users to regulate access to their private resources and can be tuned to other Online Social Networks. The main advantages of our model are its simplicity and ease of deployment. The correctness of the access control model is verified using a logical model in Answer Set Programming. The efficiency and feasibility of our mechanism is evaluated through a prototype. [ABSTRACT FROM AUTHOR]

Privacy policies are a mechanism used to inform users of the World Wide Web about the processing of their personal data. Such processing has special requirements, since personal data are regulated by data protection legislation. For example, a consent or another legal basis is typically needed. Privacy policies are documents used, among other things, to inform the data subject about processing of their personal data. These are formally represented by privacy languages. In this paper, we present a technique for constructing Layered Privacy Language policy data from web service code bases. Theoretically, we model the purposes of processing within web services by extending the privacy language with composition. We also present a formal analysis method for generating privacy policy purposes from the source code of web services. Furthermore, as a practical contribution, we present a static analysis tool that implements the theoretical solution. Finally, we report a brief case study for validating the tool [ABSTRACT FROM AUTHOR]

Despite the technical and ubiquitous nature of the Internet-of-Things (IoT), it must, at some point-of-time, make a stop to acknowledge the importance of User-eXperience (UX) in the success of any IoT endeavor. Diversity and smartness of things are not the sole factors that would impact the way users would "appreciate" IoT. Owning the technical discourse, IoT's aim is to make peoples' life easier, convenient, and safe by integrating UX into every single step of its evolution. What makes IoT unique from a UX perspective and how to rethink UX from an IoT perspective are questions raised and addressed in this paper paving the way for a win-win UX and IoT research-roadmap. [ABSTRACT FROM AUTHOR]

While the Internet has been profoundly changing how people participate in politics and is reshaping their political expressions in authoritarian countries, its exact effects remain ambiguous. This paper studies how Internet access and use influence public trust in local officials in authoritarian China. By using two-period panel survey data and combining difference-in-difference with propensity score matching, we find that, compared to Internet non-users, the Internet new-users show significantly lower political trust in local government officials. Further empirical evidence indicates that its causal mechanism primarily lies in the macro-structure of urban-rural heterogeneity and the micro-process of people's changing cognition and raised expectations. Our results substantially strengthen the argument that the diffusing power of the Internet is weakening the political legitimacy of authoritarian regimes like China. [ABSTRACT FROM AUTHOR]

Abstract Web users are flooded with information on the internet, and they feel overwhelmed by the different choices they have to make online daily. Recommender systems come to their rescue by suggesting products best aligned with their interests. To achieve this, traditional recommenders transfer users' personal data from the client to the server and dig for information about the user's interests and tastes. Moving data to the cloud violates the user confidentiality requirement and poses severe threats to user privacy and security. Moreover, with the tremendous increase in data size, it is no longer possible to collect and process massive data in the cloud. With the emergence of federated learning, numerous innovative recommender models are devised to solve these issues. In these models, the user data never leaves the client-side, and only the inferred results are sent back to the server for aggregating and updating the master model. Hence, the federated recommenders preserve user privacy and save the hassle of transferring enormous data to the cloud. This paper meticulously studies the recently proposed federated recommenders and classifies them based on the enhancements introduced in the prediction model, security scheme, or optimization technique. We identify the challenges faced by current federated recommenders and observe that most issues are inherently due to various aspects of federated learning, such as heterogeneous and non-IID data, malicious users, distributed framework, and non-reliable edge devices. While some emerge due to the coupling of the recommendation process in the federated paradigm. This research summarizes the current limitations, highlights the areas that need improvements, and presents future paths. In short, it paves the way for the development of robust federated recommenders that can handle the challenges of federated learning and, at the same time, generate high-quality recommendations. [ABSTRACT FROM AUTHOR]

Recently, phishing attacks have become one of the most prominent social engineering attacks faced by public internet users, governments, and businesses. In response to this threat, this paper proposes to give a complete vision to what Machine learning is, what phishers are using to trick gullible users with different types of phishing attacks techniques and based on our survey that phishing emails is the most effective on the targeted sectors and users which we are going to compare as well. Therefore, more effective phishing detection technology is needed to curb the threat of phishing emails that are growing at an alarming rate in recent years, thus will discuss the techniques of mitigation of phishing by Machine learning algorithms and technical solutions that have been proposed to mitigate the problem of phishing and valuable awareness knowledge users should be aware to detect and prevent from being duped by phishing scams. In this work, we proposed a detection model using machine learning techniques by splitting the dataset to train the detection model and validating the results using the test data , to capture inherent characteristics of the email text, and other features to be classified as phishing or non-phishing using three different data sets, After making a comparison between them, we obtained that the most number of features used the most accurate and efficient results achieved. the best ML algorithm accuracy were 0.88, 1.00, and 0.97 consecutively for boosted decision tree on the applied data sets. [ABSTRACT FROM AUTHOR]

This paper introduces a new method for the session construction problem, which is the first main step of the Web usage mining process. The proposed method defines user sessions as a set of navigation paths in the Web graph and produces a complete set of all possible maximal paths. Our new method is capable of generating navigation paths which cannot be extracted by using previous greedy approaches. Through experiments performed on real data, it is shown that when our new technique is used, it outperforms previous approaches in Web usage mining applications such as next-page prediction. Our analysis on Web user sessions exposes an important observation: Web users sessions contain navigation graphs that has small number of nodes where users branch out their navigation into multiple paths. [ABSTRACT FROM AUTHOR]

The tremendous growth of services available on the Internet makes user profiling an increasing important topic in web personalization. While profiling a user, one of the main challenges is that user preferences and interests usually change over time. It would be much appreciated, if the user profile could describe these changes. Such user profile is defined as time-aware user profile (TUP) in this paper. The TUP can delineate both dynamic user preferences and interests and evolutions of them, which illustrates that it can represent users more clearly and accurately. A TUP consists of a set of time slot-specific user profiles, which is defined as sub-time-aware user profile (sTUP). And a TUP is special for a user; then, its sTUPs separately represent preferences and interests of this user for different time slots. Constructing a TUP needs the constructions of its sTUPs. While constructing a sTUP from the user behavior logs for a time slot, in this paper, we present a novel learning model called personal service ecosystem (PSE) to delineate user preferences and interests naturally. Different from PSE, sTUP is the representation model. For a time slot, the sTUP is constructed based on the PSE recovered from the user behavior logs of this time slot. In terms of a TUP, two main approaches to analyze the evolutions of sTUPs in this TUP are further discussed. They are the evolution analysis of various margins between every two neighboring sTUPs called EA-VM and the evolution analysis of different sTUPs on raw data defined as EA-RD. EA-VM mainly focuses on the fine-grained evolution between two neighboring time slots by analyzing these various margins. It can be further divided into the clustering method and the distribution analysis approach. EA-RD chiefly pays attentions to the overall evolution of such sTUPs. By a comprehensive survey on the raw data of sTUPs, we summarize six interesting overall evolution patterns, including refugee pattern, periodic pattern, stable pattern, fluctuant pattern, emergency pattern and zombie pattern. Eventually, a systematic empirical study based on our collection data has been conducted, where smartphone users are taken as examples to illustrate our proposed models and approaches. And experimental results highlight the superiority of these approaches. [ABSTRACT FROM AUTHOR]

Along with the growth of the internet, the number of information published increased exponentially. This huge flow of information causes a problem called "information overload" which makes it harder for internet users to find key information they needed on the internet. To solve this, this paper proposes an application that helps user find trending news of their query/interest easily. Some challenges include how to determining the trending subtopic, how to extract only the content of each webpage, and how to present the data to user. Therefore, three core modules are used in this study, which are clustering, extraction, and presentation. Several methods are tested in this study, including naïve, manual thresholding, and heuristic clustering method. The result shows that hierarchical clustering using tf–idf word weighting, cosine similarity as distance measure and heuristically terminated using elbow point analysis achieves the best result at 50.84% Acc and 61.96% NMI. One challenge commonly faced by extraction algorithm is the tendency to have lower effectivity over time. In this paper, extraction algorithm using a prior-known subject/keyword to help the content extraction process is used. Second stage of noise removal process is also introduced to further remove noise that exists within the content block. The evaluation result shows improved score of 7.48%. The final application was able to receive score of 4.18 of 5 for its helpfulness and 4.35 of 5 for its effectiveness by respondents; showing that the proposed application could really help users to find information and help to solve information overload problem. [ABSTRACT FROM AUTHOR]

Due to high and unpredictable connection delays, privacy gaps, and traffic load of networks connecting cloud computing to end users in many of the Internet of Things (IoT)-based services, some challenges have been created in cloud computing efficiency. Hence, fog computing has been proposed as a solution in order to bring the cloud service closer to the existing things in the ecosystem. Integrating IoT with fog computing is associated with many challenges, including the resource discovery process. In one sense, sensors, devices, and things are the resources in the IoT ecosystem that searching for them regarding the quality of the search and selection can be one of the challenges in the resource discovery process. In the present paper, the hidden Markov chain learning method has been used to cope with this challenge in the IoT ecosystem integrated with the fog computing, to determine the probability of the need for each thing or resource in the near future with the aim of reducing latency and increasing the network use. The simulation in this work has been performed in the Cloudsim platform, and the considered parameters in the proposed method have been compared with TOPSIS, VIKOR and SAW methods. [ABSTRACT FROM AUTHOR]

Cloud computing has emerged as a new cost-efficient technology that provides on-demand resources over the Internet for users who pay only for their actual use. Load balancing plays an important role in cloud computing; it schedules the tasks on the virtual machines effectively to ensure cost-efficient execution of users tasks and optimal utilization of cloud resources. Because load balancing is a NP-hard optimization problem, much effort has been directed toward proposing fast algorithms that approximate the optimal solution. This paper deals with this problem and proposes new load balancing algorithms that meet requirements of cloud users and providers by reducing the makespan and improving resource utilization. For this, we modeled load balancing as a bin-stretching problem. By adopting the Worst-Fit heuristic to the bin-stretching problem, we propose a new load balancing algorithm called Worst-Fit-Based Load balancing algorithm (WFBLBA). Furthermore, by investigating the Decreasing Worst-Fit heuristic, we propose a decreasing variant of load balancing algorithm (WFDBLBA). Experimental evaluation using CloudSim simulator show that our algorithms not only outperform compared heuristics in terms of makespan, resource utilization and waiting time, but also cope better with high machine heterogeneity than compared ones. [ABSTRACT FROM AUTHOR]

Information-Centric Networking (ICN) emerges as a promising architecture for the future Internet to keep up with the tremendous growth of Internet users and the explosive increase of mobile data traffic. ICN can be coupled with Disruption Tolerant Networking (DTN) to overcome the limitations of the current Internet architecture in some challenging environments where we cannot establish a network connection, and also lighten the load on the Internet backhaul. Unfortunately, resources constraints, such as energy and buffer space, are among the most serious issues in Mobile Information-Centric Disruption Tolerant Networking (MICDTN). In this paper, we introduce an opportunistic DTN-based forwarding mechanism of Interest and Data for ICN, named Fast-MICDTN. The propagation of the content have been modeled using Markov-Decision Process (MDP). Hence, we characterize the living period in transit of the content to reach the maximum possible number of mobiles. Through extensive simulations based on synthetic mobility model and real-world traces, we prove the efficiency of the new approach in sharing content within a large number of users under a low cost. [ABSTRACT FROM AUTHOR]

With the rapid development of mobile internet, location-based services have been drawing great attention of both industry and academia. Given a dataset of location-based services, the search engine needs to recommend a proper set of services to a mobile internet user based on both query keywords and one's location. To realize great scalability, flexibility and cost saving, the data owners tend to outsource the dataset to the cloud untrusted. In this paper, we provide a secure and efficient service retrieval scheme in cloud computing. Specifically, a novel encrypted index structure is designed by taking both service texts and locations into consideration. Moreover, a depth-first search algorithm is created for the index and it can efficiently return query results. Theoretical analysis and simulation results illustrate security and efficiency of our scheme. [ABSTRACT FROM AUTHOR]

Malicious Uniform Resource Locators (URLs) embedded in emails or Twitter posts have been used as weapons for luring susceptible Internet users into executing malicious content leading to compromised systems, scams, and a multitude of cyber-attacks. These attacks can potentially might cause damages ranging from fraud to massive data breaches resulting in huge financial losses. This paper proposes a hybrid deep-learning approach named URLdeepDetect for time-of-click URL analysis and classification to detect malicious URLs. URLdeepDetect analyzes semantic and lexical features of a URL by applying various techniques, including semantic vector models and URL encryption to determine a given URL as either malicious or benign. URLdeepDetect uses supervised and unsupervised mechanisms in the form of LSTM (Long Short-Term Memory) and k-means clustering for URL classification. URLdeepDetect achieves accuracy of 98.3% and 99.7% with LSTM and k-means clustering, respectively. [ABSTRACT FROM AUTHOR]

Mobile Internet allows users to connect to the Internet anytime and anywhere. It has penetrated into every corner of human social life, and has had a tremendous impact, which has aroused great attention from all over the world. This paper introduces the concept of mobile Internet architecture, reference model and basic knowledge of technical characteristics; and then expounds the development status of key technologies of mobile Internet, combining the development trend of the world of information and network technology, the mobile Internet technology development are forecasted. Mobile Internet is a combination of mobile communication and traditional Internet which is the IT field at present and in the future for a long period of time. The rapid growth of data show that the global mobile Internet is still in the primary stage, there is still many problems to be solved and is not clear. This paper first introduces the basic concept of mobile Internet, including the definition, function the characteristics and architecture; basic research system given in the mobile Internet, discusses its components, including mobile terminals, access network, application service and the security and privacy aspects of the research status, existing problems and solutions. Finally, we discuss the research and development trend of mobile Internet in the future. The experimental results validated the performance of the proposed method, it outperforms compared with the other state-of-the-art models. [ABSTRACT FROM AUTHOR]

Predicting the browsing behavior of the user on the web has gained significant importance, as it improves the productivity of the website owners and also raises the interest of web users. The Markov model has been used immensely for user's web navigation prediction. To enhance the coverage and accuracy of the Markov model, higher order Markov models are integrated with lower order models. However, this integration results in large state-space complexity. To reduce the state-space complexity, this paper proposes a novel technique, namely Pruned all-Kth modified Markov model (PKM3). PKM3 eliminates the irrelevant states from a higher order model, which have a negligible contribution toward prediction. The proposed model is evaluated on four standard weblogs: BMS, MSWEB, CTI and MSNBC. PKM3 performance was optimal for the website in which pages were closely placed and share high interlinking. This pruning-based optimal model achieves a significant reduction in state-space complexity while maintaining comparable accuracy. [ABSTRACT FROM AUTHOR]

A sitemap represents an explicit specification of the design concept and knowledge organization of a website and is therefore considered as the website's basic ontology. It not only presents the main usage flows for users, but also hierarchically organizes concepts of the website. Typically, sitemaps are defined by webmasters in the very early stages of the website design. However, during their life websites significantly change their structure, their content and their possible navigation paths. Even if this is not the case, webmasters can fail to either define sitemaps that reflect the actual website content or, vice versa, to define the actual organization of pages and links which do not reflect the intended organization of the content coded in the sitemaps. In this paper we propose an approach which automatically generates sitemaps. Contrary to other approaches proposed in the literature, which mainly generate sitemaps from the textual content of the pages, in this work sitemaps are generated by analyzing the Web graph of a website. This allows us to: i) automatically generate a sitemap on the basis of possible navigation paths, ii) compare the generated sitemaps with either the sitemap provided by the Web designer or with the intended sitemap of the website and, consequently, iii) plan possible website re-organization. The solution we propose is based on closed frequent sequence extraction and only concentrates on hyperlinks organized in "Web lists", which are logical lists embedded in the pages. These "Web lists" are typically used for supporting users in Web site navigation and they include menus, navbars and content tables. Experiments performed on three real datasets show that the extracted sitemaps are much more similar to those defined by website curators than those obtained by competitor algorithms. [ABSTRACT FROM AUTHOR]

Recommender systems were first conceived to provide suggestions of interesting items to users. The evolution of such systems provided an understanding that a recommender system is currently used to diverse objectives. One of the current challenges in the field is to have approaches of recommendation that go beyond accuracy metrics. Since it is a very recent interest of the community, this review, also characterized as an exploratory search, provides an overview of the techniques in the area that tries to look beyond accuracy. More specifically, one of the characteristics that would provide such evolution to these systems is the adaptation. This review is then performed to find the existence and characteristics of such approaches. Of the total 438 papers returned in the submission of the search string, 57 papers were analyzed after two filtering processes. The papers have shown that the area is little explored and one of the reasons is the challenge to validate non-accuracy characteristics in such approaches. [ABSTRACT FROM AUTHOR]

Many famous online social networks, e.g., Facebook and Twitter, have achieved great success in the last several years. Users in these online social networks can establish various connections via both social links and shared attribute information. Discovering groups of users who are strongly connected internally is defined as the community detection problem. Community detection problem is very important for online social networks and has extensive applications in various social services. Meanwhile, besides these popular social networks, a large number of new social networks offering specific services also spring up in recent years. Community detection can be even more important for new networks as high quality community detection results enable new networks to provide better services, which can help attract more users effectively. In this paper, we will study the community detection problem for new networks, which is formally defined as the 'New Network Community Detection' problem. New network community detection problem is very challenging to solve for the reason that information in new networks can be too sparse to calculate effective similarity scores among users, which is crucial in community detection. However, we notice that, nowadays, users usually join multiple social networks simultaneously and those who are involved in a new network may have been using other well-developed social networks for a long time. With full considerations of network difference issues, we propose to propagate useful information from other well-established networks to the new network with efficient information propagation models to overcome the shortage of information problem. An effective and efficient method, Cat (Cold stArT community detector), is proposed in this paper to detect communities for new networks using information from multiple heterogeneous social networks simultaneously. Extensive experiments conducted on real-world heterogeneous online social networks demonstrate that Cat can address the new network community detection problem effectively. [ABSTRACT FROM AUTHOR]

Although the Internet has had a profound influence on China's economy and society, the empirical research pertaining to political consequences of the Internet is highly sparse. This paper investigates the potential relationship between Internet use and trust in government of Chinese netizens. We find robust evidence of a significant positive effect of Internet use on trust in government of netizens in China. The beneficial impact of Internet use on trust in government is found to be heterogeneous across genders and age groups. We also find that the source of information plays a role in explaining Chinese netizens' attitudes towards their governments. In addition, we find that the key mechanisms through which Internet use affects trust in government are the changes in netizens' appraisal of government performance, deference to government authority and internal efficacy. [ABSTRACT FROM AUTHOR]

Cookies are used for tracking user sessions in Web servers. Though, the security vulnerability of cookies may cause session being hijacked. To resist attacks, Dacosta et al. proposed one-time cookies (OTC) protocol. Unfortunately, one primary weakness is its availability relying on time synchronization between two machines, while the other is using a fixed session key to generate OTC during session period, turning possible adversaries to crack the key. Motivated by these shortcomings, a novel OTC protocol based on a one-time password (OTP) is proposed in the paper. The protocol adopts the OTP algorithm based on a hash chain to avoid time synchronization problems and generate a dynamic key for improving the security of OTC. For efficiency, we also enhanced the OTP algorithm. Security analysis and experimental results show that the proposed OTC protocol is promising to deliver high security and minimal burden on performance. [ABSTRACT FROM AUTHOR]

Spreading rumors on the Internet has become increasingly pervasive due to the proliferation of online social media. This paper investigates how rumors are amplified by a group of users who share similar interests or views, dubbed as an echo chamber. To this end, we identify and analyze 'rumor' echo chambers, each of which is a group of users who have participated in propagating common rumors. By collecting and analyzing 125 recent rumors from six popular fact-checking sites, and their associated 289,202 tweets/retweets generated by 176,362 users, we find that the rumors that are spread by rumor echo chamber members tend to be more viral and quickly propagated than those that are not spread by echo chamber members. We propose the notion of an echo chamber network that represents relations among rumor echo chambers. By identifying the hub rumor echo chambers (in terms of connectivity to other rumor echo chambers) in the echo chamber network, we show that the top 10% of hub rumor echo chambers contribute to propagation of 24% rumors by eliciting more than 36% of retweets, implying that core rumor echo chambers significantly contribute to rumor spreads. [ABSTRACT FROM AUTHOR]

Commercial web pages contain several blocks of information. Apart from the core content blocks, there exist subsidiary blocks like privacy notifications, advertisements, navigations and copyrights. Such blocks are called as noise information blocks. The information present in the noise blocks can deteriorate the performance of Information Retrieval. Eliminating these noises becomes a great challenge. This paper aims to extract the vital information from the web pages by eliminating noise. Once the content has been extracted, it is cleaned and presented in a standard format. We propose a system wherein content from the web pages are extracted using an unsupervised technique. This involves the utilization of the web page segmentation technique wherein the webpage is partitioned into incoherent visual blocks. Statistical methods of clustering are used to classify the visual blocks based on their features. The content that contributes to the required search is filtered from the web page and the user is presented with a webpage clean from noise. Our approach concentrates on eliminating the Crucial Noises, Auxiliary Content, Noise content based on block importance. Importance of every block is computed using the Hybrid hash algorithm. With respect to the threshold value, appropriate blocks are selected using the Enhanced-Sketching algorithm, which makes the Webpage suitable for effective Information Retrieval. [ABSTRACT FROM AUTHOR]

In this advanced era, where we have high-speed connectivity, it is very imperative to insulate medical data from forgery and fraud. With the regular increment in the number of internet users, it is challenging to transmit the beefy medical data. This (medical data) is always reused for different diagnosis purposes, so the information of the medical images need to be protected. This paper introduces a new scheme to ensure the safety of the medical data, which includes the use of a chaotic map on the fractional discrete cosine transform (FrDCT) coefficients of the medical data/images. The imperative FrDCT provides a high degree of freedom for the encryption of the medical images. The algorithm consists of two significant steps, i.e., application of FrDCT on an image and after that chaotic map on FrDCT coefficients. The proposed algorithm discusses the benefits of FrDCT over fractional Fourier transform (FRFT) concerning fractional order α. The key sensitivity and space of the proposed algorithm for different medical images inspire us to make a platform for other researchers to work in this area. Experiments are conducted to study different parameters and challenges. The proposed method has been compared with state-of-the-art techniques. The results suggest that our technique outperforms many other state-of-the-art techniques. Graphical Abstract Overview of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Social websites available over the Internet help users in predicting future events. User preferences over the web are in the form of blogs, textual content, public forum discussion and social media. However, reading informal discussions impact the users in the global market. In this paper we have proposed an SVM based classification method which assists the users to make informed transactions. Classifier makes use of the information available in the social media to make correct decision in the market. Our approach is compared with the traditional human based prediction approach and the results indicate that the classifier based prediction approach produces higher accuracy in terms of prediction compared to the traditional approach. Market window size is set to 30, 60 and 90 days respectively. [ABSTRACT FROM AUTHOR]

In this paper, we have investigated the outage capacity of secondary user for opportunistic spectrum sharing under the joint peak and average received power constraints for Rayleigh fading environment. Under this communication scenario, on detecting the licensed primary user inactive, the secondary unlicensed users transmit data/information in the licensed frequency band such that no or minimum interference may be experienced by the primary user. The soft sensing information (SSI) and secondary user's channel state information is used to obtain the closed form expressions for the ergodic and outage capacity using truncated channel inversion with fixed rate technique under the joint peak and average received power constraints. Numerically simulated results are provided to demonstrate the improvement in outage capacity of secondary user under the proposed spectrum sharing scheme. Moreover, the proposed scheme is also compared with other conventional spectrum sharing schemes to illustrate the benefits of SSI and received power constraints on the outage capacity of secondary user. [ABSTRACT FROM AUTHOR]

With the tremendous increment of traffic in the next generation mobile networks, device to device (D2D) communication is proposed to relieve the traffic burden of the base station and improve the overall network capacity. It supports direct communications between devices and could reuse the resources of cellular users (CUs). Despite the advantages, D2D communications bring great challenges in interference management. In this paper, we study the power control and channel allocation problems in three scenarios: (1) one CU and one D2D pair; (2) one CU and multiple D2D pairs; (3) multiple CUs and multiple D2D pairs. The goal is to coordinate the mutual interferences and maximize the overall network capacity. We derive sufficient conditions to guarantee the efficiency of D2D communications in scenarios with one CU and one D2D pair. We propose the bio-inspired PSO-P power control algorithm for the scenarios with one CU and multiple D2D pairs, and the PSO-CP algorithm for the scenarios with multiple CUs and multiple D2D pairs to jointly assign channels and powers. Simulation results show that the proposed algorithms are efficient in improving the overall network capacity. [ABSTRACT FROM AUTHOR]

Although social media platforms serve diverse purposes, from social and professional networking to photo sharing and blogging, people frequently use them to share the thoughts and opinions and most importantly, their interests (e.g., politics, economy, sports). Understanding the interests of social media users is key to many applications that need to characterize them to recommend some services and find other individuals with similar interests. In this paper, we propose two approaches to the automatic determination of the interests of social media users. The first, that we named FRISK, is an unsupervised multilingual approach that determines the interests of a user from the explicit meaning of the words that occur in the user's posts. The second, that we termed ASCERTAIN, is a supervised approach that resorts to the hidden dimensions of the words that several studies indicated to be capable of revealing some of the psychological processes and personality traits of a person. In our evaluation, that we performed on two datasets obtained from Twitter, we show that FRISK is capable of inferring the interests in a multilingual context with good accuracy and that the psychological dimensions used by ASCERTAIN are also good predictors of a user's interests. [ABSTRACT FROM AUTHOR]

Cloud-based databases often decouple database instances from physical storage to provide reliability and high availability to users. This design can robustly handle a single point of failure but needs substantial effort to attain good performance. In this paper, we analyze the decoupled architecture and present important optimization issues that we found while implementing this design. For implementing the prototyped system, we decouple MariaDB into a database instance and storage managing both transaction log and data pages. We also explore design considerations for transaction processing performance in the front-end side and efficient way of guaranteeing data durability in the back-end side. We evaluate the prototyped system using standard database benchmarks, and experimental results show that adopting the decoupled architecture for cloud-based DBMSs needs to address numerous optimization issues for achieving better performance. [ABSTRACT FROM AUTHOR]

This paper aims to examine the factors influencing older Chinese's social networking site (SNS) use and to help older people live a better life in this technology-oriented world. A proposed model composed of eight hypotheses was built according to Diffusion of Innovation Theory and was tested via 726 valid data points. The empirical results show that technical compatibility and relative advantages as technical factors and Internet self-efficacy as a personal cognitive factor promote older people's SNS use. Technical complexity as a technical factor and Internet anxiety as a personal cognitive factor hinder older people's SNS use. Social support as a facilitating condition alleviates the negative effects of technical complexity and Internet anxiety and enhances the positive effect of Internet self-efficacy. Gender differences and health status differences are significant. Older men are more likely to use SNSs than older women; older adults in poor physical health condition are more likely to use SNSs compared with those in good health. Compared with older people in Western countries, older Chinese's SNS use does show some differences: 1) gender differences are significant with males having a stronger intention to use SNSs; 2) older people face severe digital divide because technical factors show stronger power in predicting their SNS use; 3) older people need social support to assist their SNS use, but their social support mainly comes from age peers rather than young people. [ABSTRACT FROM AUTHOR]

Social networks facilitate communication among Internet users while generating large volumes of online short-text conversations every day. This leads to a huge number of free-style asynchronous conversations where multiple users are involved and multiple topics are discussed at the same time in the same place, e.g., an instant group chat in WeChat. Here emerges an interesting problem: as a result of a large number of users and topics, the conversation structure may get into a mess, which often interferes with the acquisition of messages users are interested in. For example, when a user enters a conversation, (s)he usually does not want to read all the historical messages, but just hope to get the messages that are the most relevant to some messages (s)he cares about. Therefore, it is an essential task to understand the logical correlations among messages, which benefits text mining, natural language processing, and web intelligence techniques. In this paper, we focus on "reply-to" relations, such as Q&A between messages in group chats. At first, a model called LSTM-RT is presented to predict the "reply-to" relations between messages, which is based on deep bidirectional LSTM networks. Then, three versions of the LSTM-RT model are proposed. In detail, the first version is based on a non-siamese architecture, which processes ordered message pairs; The other two versions are end-to-end models, which are based on the word level and the sentence level, respectively. Finally, experimental results conducted on two real-world group chat data sets demonstrate the effectiveness of the proposed model. [ABSTRACT FROM AUTHOR]

Anonymity protocols are employed to establish encrypted tunnels to protect the privacy of Internet users from traffic analysis attacks. However, the attackers strive to infer some traffic patterns’ characteristics (e.g. packet directions, packet sizes, inter-packet timing, etc.) in order to expose the identities of Internet users and their activities. A recent and popular traffic analysis attack is called website fingerprinting which reveals the identity of websites visited by target users. Existing work in the literature studied the website fingerprinting attack using a single web browser, namely Firefox. In this paper we propose a unified traffic analysis attack model composed of a sequence of phases that demonstrate the efficiency of website fingerprinting attack using popular web browsers under Tor (The Onion Router). In addition, we reveal the main factors that affect the accuracy of website fingerprinting attack over Tor anonymous system and using different browsers. To the best of our knowledge, no previous study uncovered such factors by deploying real-world traffic analysis attack utilizing the top five web browsers. The outcomes of the research are very relevant to Internet users (individuals/companies/governments) since they allow to assess to which extent their privacy is preserved in presence of traffic analysis attacks, in particular, website fingerprinting over different browsers. A recommendation for future research direction regarding the investigation of website fingerprinting over different scenarios is also provided. [ABSTRACT FROM AUTHOR]

Considering the existence of a wide range of web services and their diversity, selecting a service with the best performance among similar services remains a challenge for the demanding users. However, sometimes it is necessary that web services invoke each other and become combined to create a more complex composite web service. Evaluation of the trust degree based on the trust of individual services forming composite ones can be an effective factor for choosing the best combination of web services. In this paper, by using the fundamental concepts of belief combination and transferability in transferable belief model (TBM), the ratings of a customer for different interactions are combined to constitute the direct trust value. This creates a requirement to take the uncertainty of beliefs into account as well. In direct trust evaluation, the time of ratings determines the impact of the ratings in the final combination. By using the generalized Bayesian theorem in TBM, the trust of the composite services is modeled through time by considering the structure of the individual components within the service. Subsequently, it is shown that the proposed model provides a flexible mechanism for composite trust evaluation, it provides more accurate calculations compared to other existing models, and it is capable of taking into account more complex parameters in trust calculations such as the consistency of opinions. [ABSTRACT FROM AUTHOR]

Every time an Internet user downloads a video, shares a picture, or sends an email, his/her device addresses a data center and often several of them. These complex systems feed the web and all Internet applications with their computing power and information storage, but they are very energy hungry. The energy consumed by Information and Communication Technology (ICT) infrastructures is currently more than 4% of the worldwide consumption and it is expected to double in the next few years. Data centers and communication networks are responsible for a large portion of the ICT energy consumption and this has stimulated in the last years a research effort to reduce or mitigate their environmental impact. Most of the approaches proposed tackle the problem by separately optimizing the power consumption of the servers in data centers and of the network. However, the Cloud computing infrastructure of most providers, which includes traditional telcos that are extending their offer, is rapidly evolving toward geographically distributed data centers strongly integrated with the network interconnecting them. Distributed data centers do not only bring services closer to users with better quality, but also provide opportunities to improve energy efficiency exploiting the variation of prices in different time zones, the locally generated green energy, and the storage systems that are becoming popular in energy networks. In this paper, we propose an energy aware joint management framework for geo-distributed data centers and their interconnection network. The model is based on virtual machine migration and formulated using mixed integer linear programming. It can be solved using state-of-the art solvers such as CPLEX in reasonable time. The proposed approach covers various aspects of Cloud computing systems. Alongside, it jointly manages the use of green and brown energies using energy storage technologies. The obtained results show that significant energy cost savings can be achieved compared to a baseline strategy, in which data centers do not collaborate to reduce energy and do not use the power coming from renewable resources. [ABSTRACT FROM AUTHOR]

Cognitive radio technology provides opportunistic wireless spectrums access for the secondary users (SUs) while primary users (PUs) are dormant. By emergence of a PU in the cognitive radio networks, SUs are required to vacant the channel by using spectrum handoff approaches to avoid any collision. Providing an efficient and flexible coordination protocol for spectrum handoff is a challenging step. In this paper, we propose a novel proactive spectrum handoff protocol called multiple-single rendezvous (MSRV) protocol which uses multiple rendezvous (MRV) and single rendezvous (SRV) coordination policy to provide a higher throughput and context adaptability. MRV coordination policy has the advantage of negotiating in different channels simultaneously and avoid single control channel congestion. On the other hand, in order to share the PUs’ channel usage history information between the SUs for predicting channels’ availability, it is necessary to utilize a SRV coordination scheme. The proposed MSRV protocol improves the average throughput of the SUs and decreases the average service time comparing to the other existing proactive spectrum handoff protocols. Moreover, MSRV protocol gives priority to the handoff SUs which decreases service times. It is an crucial property for the delay sensitive applications. Our proposed protocol achieves 20% throughput and 13.7% average service time improvement in 1.25 and 10 (pkts/s) PUs’ traffic load compared to the SRV proactive spectrum handoff protocol proposed by Mehrnoush et al., respectively. [ABSTRACT FROM AUTHOR]

Nowadays, search engines play a gateway role for users to access their needed information in the Web. However, malicious users can also use them to facilitate their attacks by submitting excessive amounts of bot-generated queries, called spam queries. In this paper, we propose a novel semi-supervised method which can effectively detect spam queries in a practical manner. We first train a model to characterize normal and malicious users, using the linguistic properties of queries as well as the behavioral characteristics of users and IP addresses. Then, we use the trained model to predict the label of arriving requests with a fast and efficient algorithm which works based on the stream clustering approach. The results of our evaluation with the real log of a local search engine show that the proposed algorithm yields an accuracy of about %94, while incurring a low response-time and memory overhead. [ABSTRACT FROM AUTHOR]

The large diffusion of Online Social Networks (OSNs) has influenced the way people interact with each other. OSNs present several drawbacks, one of the most important is the problem of privacy disclosures. Distributed Online Social Networks (DOSNs) have been proposed as a valid alternative solution to solve this problem. DOSNs are Online Social Networks implemented on a distributed platform, such as a P2P system or a mobile network. However, the decentralization of the control presents several challenges, one of the main ones is guaranteeing data availability without relying on a central server. To this aim, users’ data allocation strategies have to be defined and this requires the knowledge of both structural and temporal characteristics of ego networks which is a difficult task due to the lack of real datasets limiting the research in this field. The goal of this paper is the study of the behaviour of users in a real social network in order to define proper strategies to allocate the users’ data on the DOSN nodes. In particular, we present an analysis of the temporal affinity and the structure of communities and their evolution over the time by using a real Facebook dataset. [ABSTRACT FROM AUTHOR]

Web objects are the entities retrieved from websites by users to compose the web pages. Therefore, exploring the relationships among web objects has theoretical and practical significance for many important applications, such as content recommendation, web page classification, and network security. In this paper, we propose a graph model named Bipartite Request Dependency Graph (BRDG) to investigate the relationships among web objects. To build the BRDG from massive network traffic data, we design and implement a parallel algorithm by leveraging the MapReduce programming model. Based on the study of a number of BRDGs derived from real wireless network traffic datasets, we find that the BRDG is large, sparse and complex, implying that it is very hard to derive the structural characteristics of the BRDG. Towards this end, we propose a co-clustering algorithm to decompose and extract coherent co-clusters from the BRDG. The co-clustering results of the experimental dataset reveal a number of interesting and interpretable connectivity structural patterns among web objects, which are useful for more comprehensive understanding of web page architecture and provide valuable data for e-commerce, social networking, search engine, etc. [ABSTRACT FROM AUTHOR]

The pilot contamination caused by sharing the non-orthogonal pilots among users is considered to be a bottleneck of the massive multi-input multi-output (MIMO) systems. This paper proposes a pilot scheduling scheme based on the degradation to address this problem. Through computing the degradation of the users, the proposed scheduling assigns the optimal pilot sequence to the user who suffers from the greatest degradation in a greedy way. Moreover, the proposed scheme is further optimized with an extra set of orthogonal pilot sequences, which is called pilot scheduling scheme based on user grouping. Simulation results show that the target cell's achievable sum rate of the proposed scheme is much higher than the random pilot scheduling (RPS) and the smart pilot assignment (SPA) schemes; also, our scheme can reduce the impact of shadowing fading on the target cell's achievable sum rate effectively. [ABSTRACT FROM AUTHOR]

An ensemble of neural networks has been proved to be an effective machine learning framework. However, very limited studies in the current literature examined the neural network ensemble for online regression; furthermore, these methods were combination of online individual models and did not consider the ensemble diversity. In this paper, a novel online sequential learning algorithm for neural network ensembles for online regression is proposed. The algorithm is built upon the decorrelated neural network ensembles (DNNE) and thus referred to as Online-DNNE; so it uses single-hidden layer feed-forward neural networks with random hidden nodes' parameters as ensemble components and introduces negative correlation learning to train base models simultaneously in a cooperative manner which can effectively maintain the ensemble diversity. The Online-DNNE only learns the newly arrived data, and the computation complexity is thus reduced. The results of the experiments with benchmarks show the effectiveness and significant advantages of the proposed approach. [ABSTRACT FROM AUTHOR]

Most East Asian countries have policies to promote digital inclusion, especially among the children of low-income families and persons with disabilities, for the purposes of social development. However, such efforts and their impact among older people have been rather limited. In Hong Kong, only 7 % of those aged 65 or above were Internet users in 2008, compared with 66.7 % of the general public. Providing computer and Internet training programmes to older people is thus crucial in promoting digital inclusion. This paper discusses the experience and results of training older people to use both a computer and the Internet. Based on findings from an earlier qualitative study, we hypothesise that acquiring knowledge of computers and Internet use will have an empowering effect for older people in terms of increasing self-efficacy and enhancing communication with friends and family members. This paper presents two studies of such empowerment among older people in two non-governmental organisations (NGOs). While older people with little computer knowledge showed some effects of empowerment, at the same time, they also became more aware of the marginalisation and exclusion effects of their limited knowledge of computer and Internet usage. However, among users with some prior experience of using a computer, further training failed to show additional positive empowerment results. The results of the two studies call for a re-examination of the training process and experience of older people as well as integration of such training with other strategies to achieve better results in promoting their social inclusion in the information society. [ABSTRACT FROM AUTHOR]

This paper proposes a novel architecture of the ID (IDentifier) layer for Internet of Things, which is embedded in the network level instead of traditional overlay solutions. Our contribution characterizes development of human-readable, hierarchical ID-based unified addressing for both objects and services, corresponding to their locations. In this way, users gain easier access to the IoT resources and native support for multicast. Furthermore, we take advantage of the Networking Named Content approach to specify rules for ID-based data transfer. The network nodes have capabilities to cache forwarded data for handling future requests, what may decrease network overload and facilitate cooperation between applications and sensors which periodically move into sleep mode for saving energy. ID-based routing offers decoupling of identification of objects/services from their location. Awareness of forwarded IoT data together with hierarchical distribution of the network makes feasible the local management of users and objects, that is, the essential IoT processes such as object/service registration, publication, searching and resolution can be managed locally in the network node. This offers high flexibility and manageability and improves response of the system when the number of handled things scales. The paper presents the architecture of the proposed solution for ID layer focusing on modules and mechanisms of the ID network node, as well as detailed description of registration, resolution and forwarding processes. Furthermore, the implementation of the system, performed on Linux-based routers, was tested to confirm the correctness of ID layer processes. The tests show that the performance of ID network node is not worse than the performance of a classical IP Linux router running on an identical physical device. [ABSTRACT FROM AUTHOR]

Twitter is one of the most popular applications in the current Internet with more than 500 M registered users across the world. In this paper, we conduct a comprehensive analysis to understand the geographical characteristics of Twitter using cross-community mining techniques. Specifically, we study the locality level shown by the three main elements of Twitter, namely users, relationships and information flow. For this purpose, we rely on a dataset including the geolocation information of more than 17, 100 and 3.5 M users, relationships and tweets, respectively. Our main findings are: (1) most of the Twitter users perform their activity from an area of at most few hundred kms covering few cities within a unique country; (2) the location (i.e., country), and in particular factors such as language or Twitter popularity within a country, dictates the level of locality in the relationships of users and Twitter conversations originated in that country. The combination of these factors reveals the presence of four types of country locality profiles that we carefully analyze and compare in the paper. [ABSTRACT FROM AUTHOR]

The part of the Internet of Things composed of devices that directly interact with users has grown considerably in the past years. With new smartphones, tablets and other Internet-enabled devices that appear on the market, this trend is still increasing. However, existing application development processes and tools, designed for single device applications, do not allow developers to fully and efficiently address this opportunity. Applications are developed for a particular type of devices or a particular programming platform. This limits the number of potential users and makes it difficult to seamlessly use an application on multiple devices owned by users. To take full advantage of the Internet of Things, applications should be able to run on any device-they should be ubiquitous. In this paper, we present a concept of Device-Independent Architecture, which provides separation of applications from devices and facilitates development of device-independent applications. Additionally, the separation introduced by the Device-Independent Architecture enables implementation of multi-device scenarios where a single application employs multiple devices at the same time. The experiment described in the paper proves that such device-independent applications indeed may be used on any suitable device-they have a chance to become ubiquitous. [ABSTRACT FROM AUTHOR]

Data centers have become essential to modern society by catering to increasing number of Internet users and technologies. This results in significant challenges in terms of escalating energy consumption. Research on green initiatives that reduce energy consumption while maintaining performance levels is exigent for data centers. However, energy efficiency and resource utilization are conflicting in general. Thus, it is imperative to develop an application assignment strategy that maintains a trade-off between energy and quality of service. To address this problem, a profile-based dynamic energy management framework is presented in this paper for dynamic application assignment to virtual machines (VMs). It estimates application finishing times and addresses real-time issues in application resource provisioning. The framework implements a dynamic assignment strategy by a repairing genetic algorithm (RGA), which employs realistic profiles of applications, virtual machines and physical servers. The RGA is integrated into a three-layer energy management system incorporating VM placement to derive actual energy savings. Experiments are conducted to demonstrate the effectiveness of the dynamic approach to application management. The dynamic approach produces up to 48% better energy savings than existing application assignment approaches under investigated scenarios. It also performs better than the static application management approach with 10% higher resource utilization efficiency and lower degree of imbalance. [ABSTRACT FROM AUTHOR]