Search

Keystroke dynamics is the field of computer science that exploits data derived from the way users type. It has been used in authentication systems, in the identification of user characteristics for forensic or commercial purposes, and to identify the physical and mental state of users for purposes that serve human–computer interaction. Studies of keystroke dynamics have used datasets created from volunteers recording fixed-text typing or free-text typing. Unfortunately, there are not enough keystroke dynamics datasets available on the Internet, especially from the free-text category, because they contain sensitive and personal information from the volunteers. In this work, a free-text dataset is presented, which consists of 533 logfiles, each of which contains data from 3500 keystrokes, coming from 164 volunteers. Specifically, the software developed to record user typing is described, the demographics of the volunteers who participated are given, the structure of the dataset is analyzed, and the experiments performed on the dataset justify its utility. [ABSTRACT FROM AUTHOR]

The rapid development of the smart grid affects every aspect of people's lives at all times. Among them, the identification accuracy of power metering anomalies is a serious problem that needs to be solved urgently. To further improve the identification accuracy of power metering anomalies, a user identification method for substation area anomalies is designed based on the restricted ridge regression model. First, a clustering algorithm is designed using high-frequency voltage data of users in the substation area to conduct phase separation and terminal branch identification for each user metering point in the substation area. Then establish a substation area energy conservation model equation set, and merge the end branch users based on genetic algorithm clustering results, thereby reducing the unknown quantity dimension. Then, a line resistance-constrained condition is constructed using the topological structure, and the model equation set is solved using quadratic programming based on the constrained ridge regression method to obtain the measurement point error results of users in the substation area. The experimental results show that this recognition method's clustering effect and recognition accuracy are high, with a recognition rate of 98.21%. It can be seen that this abnormal user identification method has application value. [ABSTRACT FROM AUTHOR]

Integrated energy system (IES) is an important means to achieve the goal of “carbon peaking and carbon neutrality”. However, different types of users in the system have different energy consumption behaviors, which makes the coordinated optimization and low-carbon operation of the integrated energy system more difficult. In order to give full play to the subjective initiative of users, the user behavior of the integrated energy system is modelled based on user behavior analysis, and users are classified into aggressive and conservative types by convolutional neural network (CNN). Then, the decision model of integrated energy system operator is constructed to determine the supply mode of electric heating energy, and the corresponding energy package is designed for different types of users. Finally, the effectiveness of the above models and methods is analyzed based on actual data, and the value of user classification in low-carbon operation of integrated energy systems is verified.

Keystroke dynamics is the field of computer science that exploits data derived from the way users type. It has been used in authentication systems, in the identification of user characteristics for forensic or commercial purposes, and to identify the physical and mental state of users for purposes that serve human–computer interaction. Studies of keystroke dynamics have used datasets created from volunteers recording fixed-text typing or free-text typing. Unfortunately, there are not enough keystroke dynamics datasets available on the Internet, especially from the free-text category, because they contain sensitive and personal information from the volunteers. In this work, a free-text dataset is presented, which consists of 533 logfiles, each of which contains data from 3500 keystrokes, coming from 164 volunteers. Specifically, the software developed to record user typing is described, the demographics of the volunteers who participated are given, the structure of the dataset is analyzed, and the experiments performed on the dataset justify its utility.

The rapid development of information and communication technologies and the widespread use of the Internet has made it imperative to implement advanced user authentication methods based on the analysis of behavioural biometric data. In contrast to traditional authentication techniques, such as the simple use of passwords, these new methods face the challenge of authenticating users at more complex levels, even after the initial verification. This is particularly important as it helps to address risks such as the possibility of forgery and the disclosure of personal information to unauthorised individuals. In this study, the use of keystroke dynamics has been chosen as a biometric, which is the way a user uses the keyboard. Specifically, a number of Bulgarian-speaking users have been recorded during their daily keyboard use, and then a system has been implemented which, with the help of machine learning models, recognises certain acquired or intrinsic characteristics in order to reveal part of their identity. The results show that users can be categorised using keystroke dynamics, in terms of the age group they belong to and in terms of their educational level, with high accuracy rates, which is a strong indication for the creation of applications to enhance user security and facilitate their use of Internet services.

Copyright of Journal of Shanghai Jiao Tong University (1006-2467) is the property of Journal of Shanghai Jiao Tong University Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The rapid development of information and communication technologies and the widespread use of the Internet has made it imperative to implement advanced user authentication methods based on the analysis of behavioural biometric data. In contrast to traditional authentication techniques, such as the simple use of passwords, these new methods face the challenge of authenticating users at more complex levels, even after the initial verification. This is particularly important as it helps to address risks such as the possibility of forgery and the disclosure of personal information to unauthorised individuals. In this study, the use of keystroke dynamics has been chosen as a biometric, which is the way a user uses the keyboard. Specifically, a number of Bulgarian-speaking users have been recorded during their daily keyboard use, and then a system has been implemented which, with the help of machine learning models, recognises certain acquired or intrinsic characteristics in order to reveal part of their identity. The results show that users can be categorised using keystroke dynamics, in terms of the age group they belong to and in terms of their educational level, with high accuracy rates, which is a strong indication for the creation of applications to enhance user security and facilitate their use of Internet services. [ABSTRACT FROM AUTHOR]

Understanding the distinct characteristics of unidentified Internet users is helpful in various contexts, including digital forensics, targeted advertising, and user interaction with services and systems. Keystroke dynamics (KD) enables the analysis of data derived from a user’s typing behaviour on a keyboard as one approach to obtain such information. This study conducted experiments on a developed dataset that recorded samples of typing in five different mother tongues to determine Internet users’ mother tongue. Based on only a few KD features and machine learning techniques, 82% accuracy was achieved in recognising an unknown user’s mother tongue. This research highlights the potential for KD as a reliable method for identifying the mother tongue of Internet users, with implications for various applications such as improving digital forensic investigations, targeted advertising strategies, and optimising user experiences with online services.

Power sales business stands on reasonable assessment of power users by power sales enterprises. The present assessment methods are incomprehensive and less applied. Therefore， a power user classification model is proposed based on WRSR （weighted rank-sum ratio） and Catboost algorithm. Firstly， the existing power users are graded and labeled using WRSR. Then， the Catboost algorithm is used to learn the classification law and construct a classifier. The hyperparameters of Catboost are optimized using BOA （Bayesian optimization algorithm） to improve the classification effect. Finally， the importance of each feature is analyzed according to the model， and the user features are filtered by the importance scores. The experimental results show that the method can reasonably classify power users， and the proposed classification model is superior to other machine learning models in accuracy and interpretability.

As a classification task, the detection of new malicious users in social networks has been facing the lack of datasets and labels of malicious users. With limited data, this paper proposed a method based on adaptive differential graph convolution to detect malicious users accurately. By extracting user features and social relationships in the social network, the method constructed the social network graph. After this, it calculated the similarities between node and its neighbors to prioritize the neighbors, and used the priority order to sample key neighbors. The node used adaptive weighted average to aggregate the features of key neighbors to itself, to update its features. After feature updating, by feature dimension reduction and normalization, the node got its malicious value, for malicious detection. The experiment results show that, compared to other methods, the method proposed in the paper achieves higher precision and overall accuracy on detection of new malicious users, with a satisfactory speed. Results also demonstrate that adaptive differential graph convolutional networks can effectively capture the key features of a small number of data samples. [ABSTRACT FROM AUTHOR]

Understanding the distinct characteristics of unidentified Internet users is helpful in various contexts, including digital forensics, targeted advertising, and user interaction with services and systems. Keystroke dynamics (KD) enables the analysis of data derived from a user's typing behaviour on a keyboard as one approach to obtain such information. This study conducted experiments on a developed dataset that recorded samples of typing in five different mother tongues to determine Internet users' mother tongue. Based on only a few KD features and machine learning techniques, 82% accuracy was achieved in recognising an unknown user's mother tongue. This research highlights the potential for KD as a reliable method for identifying the mother tongue of Internet users, with implications for various applications such as improving digital forensic investigations, targeted advertising strategies, and optimising user experiences with online services. [ABSTRACT FROM AUTHOR]

Copyright of Zhejiang Electric Power is the property of Zhejiang Electric Power Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Abstract This paper presents a user-centric frequency reuse scheme depending on the user classification in the cell in a homogeneous Poisson point process network. Each cell is partitioned into multiple regions delimited by the signal-to-interference ratio of the user considered. A given resource block (RB) is assigned to one user in a cell and cannot be shared by another user in this cell. The typical user only uses a fraction of the RB, and the remaining part is left unused to avoid interference with the other cells. The base stations (BS) interfering set is hence determined by the coverage probability of the typical user, and this set can be approximated by a thinned version of the original PPP. Using this, coverage probability and spectral efficiency (SE) for each user types are derived, and a BS cooperation technique is proposed to improve the SE of the cell edge user. Finally, the average network spectral efficiency and fairness among users are analyzed under the developed theoretical framework. Our results show that the user-centric frequency reuse improves the coverage probability and the SE compared to the conventional frequency reuse scheme.

Energy consumption is increasing daily, and with that comes a continuous increase in energy costs. Predicting future energy consumption and building an effective energy management system for smart homes has become essential for many industrialists to solve the problem of energy wastage. Machine learning has shown significant outcomes in the field of energy management systems. This paper presents a comprehensive predictive-learning based framework for smart home energy management systems. We propose five modules: classification, prediction, optimization, scheduling, and controllers. In the classification module, we classify the category of users and appliances by using k-means clustering and support vector machine based classification. We predict the future energy consumption and energy cost for each user category using long-term memory in the prediction module. We define objective functions for optimization and use grey wolf optimization and particle swarm optimization for scheduling appliances. For each case, we give priority to user preferences and indoor and outdoor environmental conditions. We define control rules to control the usage of appliances according to the schedule while prioritizing user preferences and minimizing energy consumption and cost. We perform experiments to evaluate the performance of our proposed methodology, and the results show that our proposed approach significantly reduces energy cost while providing an optimized solution for energy consumption that prioritizes user preferences and considers both indoor and outdoor environmental factors. [ABSTRACT FROM AUTHOR]

In the era of big data, all countries and governments generally attach importance to the opening of scientific data and actively promote data sharing.The wide opening and sharing of data poses new challenges to the whole process of data security.Based on the latest policies of international scientific data sharing and the actual situation in China, the resource and user access control system of scientific data center was put forward.Through the application of data classification, user classification and system access control strategy on the open system platform, the problem of lack of systematic scheme for resource security sharing in the data center was solved.The research results have been applied to the practical work of the National Earth Observation Scientific Data Center and achieved good results.

This paper presents a user-centric frequency reuse scheme depending on the user classification in the cell in a homogeneous Poisson point process network. Each cell is partitioned into multiple regions delimited by the signal-to-interference ratio of the user considered. A given resource block (RB) is assigned to one user in a cell and cannot be shared by another user in this cell. The typical user only uses a fraction of the RB, and the remaining part is left unused to avoid interference with the other cells. The base stations (BS) interfering set is hence determined by the coverage probability of the typical user, and this set can be approximated by a thinned version of the original PPP. Using this, coverage probability and spectral efficiency (SE) for each user types are derived, and a BS cooperation technique is proposed to improve the SE of the cell edge user. Finally, the average network spectral efficiency and fairness among users are analyzed under the developed theoretical framework. Our results show that the user-centric frequency reuse improves the coverage probability and the SE compared to the conventional frequency reuse scheme. [ABSTRACT FROM AUTHOR]

Guided by a corpus linguistics approach, in this article we present a comparative evaluation of State-of-the-Art (SotA) models, with a special focus on Transformers, to address the task of Fake News Spreaders (i.e., users that share Fake News) detection. First, we explore the reference multilingual dataset for the considered task, exploiting corpus linguistics techniques, such as chi-square test, keywords and Word Sketch. Second, we perform experiments on several models for Natural Language Processing. Third, we perform a comparative evaluation using the most recent Transformer-based models (RoBERTa, DistilBERT, BERT, XLNet, ELECTRA, Longformer) and other deep and non-deep SotA models (CNN, MultiCNN, Bayes, SVM). The CNN tested outperforms all the models tested and, to the best of our knowledge, any existing approach on the same dataset. Fourth, to better understand this result, we conduct a post-hoc analysis as an attempt to investigate the behaviour of the presented best performing black-box model. This study highlights the importance of choosing a suitable classifier given the specific task. To make an educated decision, we propose the use of corpus linguistics techniques. Our results suggest that large pre-trained deep models like Transformers are not necessarily the first choice when addressing a text classification task as the one presented in this article. All the code developed to run our tests is publicly available on GitHub. [ABSTRACT FROM AUTHOR]

In the current modern power system, extreme load peaks and valleys frequently occur due to the complicated electricity consumption behaviors. This point severely impacts the security, stability, and economy of the power system. Demand response (DR) has been proved to be one of the most effective ways to shift load to relieve the intensity of the power system. Although DR is mainly applied on the commercial and industrial loads traditionally, in recent years, the residential load has gradually attracted attentions of DR researches, especially incentive demand response (IDR) research because of its remarkable stability and flexibility in terms of load shifting. However, the difficulty of measuring the IDR adaptability and potential of a residential user according to the load curve significantly prevents the IDR from being conveniently implemented. And further, the power company is tremendously difficult to efficiently and effectively select the users with high IDR adaptabilities and potentials to participate in IDR. Therefore, to address the aforementioned issues, this paper presents a residential user classification approach based on the graded user portrait with considering the IDR adaptability and potential. Based on the portrait approach, the residential users with high IDR adaptabilities can be preliminarily selected. And then, based on the selected users, the portrait approach to delineate the users with high IDR potentials is further presented. Afterward, the achieved residential users with high adaptabilities and potentials are labeled, which are employed to train the presented variational auto encoder based deep belief network (VAE-DBN) load classification model. The experimental results show the effectiveness of the presented user portrait approaches as well as the presented load classification model. The results suggest that the presented approaches could be potential tools for power company to identify the suitable residential users for participating in the IDR tasks.

Energy consumption is increasing daily, and with that comes a continuous increase in energy costs. Predicting future energy consumption and building an effective energy management system for smart homes has become essential for many industrialists to solve the problem of energy wastage. Machine learning has shown significant outcomes in the field of energy management systems. This paper presents a comprehensive predictive-learning based framework for smart home energy management systems. We propose five modules: classification, prediction, optimization, scheduling, and controllers. In the classification module, we classify the category of users and appliances by using k-means clustering and support vector machine based classification. We predict the future energy consumption and energy cost for each user category using long-term memory in the prediction module. We define objective functions for optimization and use grey wolf optimization and particle swarm optimization for scheduling appliances. For each case, we give priority to user preferences and indoor and outdoor environmental conditions. We define control rules to control the usage of appliances according to the schedule while prioritizing user preferences and minimizing energy consumption and cost. We perform experiments to evaluate the performance of our proposed methodology, and the results show that our proposed approach significantly reduces energy cost while providing an optimized solution for energy consumption that prioritizes user preferences and considers both indoor and outdoor environmental factors.

Response efficiency, load-carrying capacity, and intelligent fault-tolerant capability are three main performance bottlenecks which restrict the overall performance of the information network, so that the application system in the network is difficult to achieve the desired satisfactory results. Aiming at the above problems, this paper proposes a new network structure based on multi-node continuous collaboration and intelligent fault tolerance. In the new structure network, the user is divided into query group and read-write group according to the difference of user's permission to operate database. The user's access load in each group is handled by a separate task processing system. The task processing system consists of gateway group, communication link, application server group, database server group, and database. The real-time data synchronization problem is solved by data synchronization mechanism between databases. The problem of fault redundancy and load balancing is solved by cluster technology. The test results show that the information network performance bottleneck is greatly eliminated, the response efficiency of heavy load interval is more than 50% higher than that of traditional structural network, and the carrying capacity and intelligent fault tolerance capacity are obviously improved. [ABSTRACT FROM AUTHOR]

We have witnessed a rapid growth over past decades in sensor data mining (SDM), which aims at extracting valuable information automatically from large repositories of moving activity data. One of the significant SDM tasks is identifying humans through their transit modes using a variety of user-tracking systems. However, to the best of our knowledge, distinguishing traces of users and understanding their behaviors are difficult tasks in most real-life cases for the following reasons: 1) activity data containing both temporal and spatial contexts are of high order and sparse; 2) living patterns are not as regular as expected, and the route choice uncertainties due to their vagueness and randomness; 3) owing to the complexity and sparseness of urban travel methods, although some deep learning-based models can produce relatively good classification results, they can still be improved by combining external information. To address these challenges, we propose a novel scenario-based deep learning method which is based on the assumption that people visit places with explicit purposes (e.g., to go to work or visit a park). We first represent semantic patterns from daily life and create various scenarios and utilize an attention neural network to embed points of trajectories by considering both semantic and geographical information. Then, we construct a Semantic Structure Neural Network (S2N2) framework to perform the end-end classification. Our S2N2 model is applied to an interesting yet challenging topic: distinguishing suspect transit behavior on a real-life data set collected by mobile devices. Although the problem is not entirely solved, the extensive evaluation presented here demonstrates that our model outperforms conventional classification methods, anomaly detection methods, and state-of-the-art sequential deep learning models, especially when trajectory semantic vectors are incorporated. We also provide statistical analysis and intuitive explanations to help interpret the characteristics of user mobility.

The enhancement of the carrying capacity of high-speed railway and the acceleration of train speed lead to the proliferation of signal interference in the communication network, which leads to the degradation of network performance and user service quality. To address the issue, we first classify the cell users in high-speed railway wireless communication environment based on Fuzzy C-Means algorithm and user mobility prediction model, and divide communicating users on trains into cell center users and edge users. Then, differential power distribution schemes are implemented for center users and edge users according to the classification results. Finally, the Max-SINR based interference alignment algorithm is used to realize the interference management. Simulation results show that the proposed algorithm fully takes into account the mobility of high-speed train users and effectively manages interference, which improves the high-speed railway wireless communication network performance significantly.

With the rapid development and popularization of new-generation technologies such as cloud computing, big data, and artificial intelligence, the construction of smart grids has become more diversified. Accurate quick reading and classification of the electricity consumption of residential users can provide a more in-depth perception of the actual power consumption of residents, which is essential to ensure the normal operation of the power system, energy management and planning. Based on the distributed architecture of cloud computing, this paper designs an improved random forest residential electricity classification method. It uses the unique out-of-bag error of random forest and combines the Drosophila algorithm to optimize the internal parameters of the random forest, thereby improving the performance of the random forest algorithm. This method uses MapReduce to train an improved random forest model on the cloud computing platform, and then uses the trained model to analyze the residential electricity consumption data set, divides all residents into 5 categories, and verifies the effectiveness of the model through experiments and feasibility. [ABSTRACT FROM AUTHOR]

At present, the research on the classification of e-commerce users is relatively mature, but with the rise of mobile social networks, the combination of social networks and e-commerce networks has become a trend and is developing rapidly. Traditional e-commerce user classification methods are not suitable for social e-commerce users. Therefore, based on the research on traditional e-commerce user classification methods, according to the characteristics of social e-commerce users, we improved data preprocessing and parameter tuning methods, and proposed a clustering method of social e-commerce users based on the K-means++ algorithm. The test on the actual data of social e-commerce users showed that the retention rates of users of various classes are significantly different, which express that the proposed method can classify social e-commerce users accurately. [ABSTRACT FROM AUTHOR]

This paper focuses on a comprehensive framework design and performance analysis for a large-scale hotspot communication scenario where there exist large number of clustered pico base stations (P-BSs) and femto BSs (F-BSs) working on millimeter-Wave (mm-Wave). For such realistic hotspots’ deployment, to exploit the coupling between user equipments (UEs) and base stations (BSs), we model the geographical centers of UE hotspots as independent PPP around which UEs, P-BSs and F-BSs are scattered and form independent and non-homogeneous Poisson cluster processes (PCPs). However, with clustered P-BSs and F-BSs, the UEs would experience the severer intra-cluster interference so that the performance is degraded greatly. To overcome this problem in clustered networks, an effective interference management scheme is proposed by using the ratio of the first and second nearest distances of a typical UE from P-BSs so that a typical UE can be classified as either cluster-center UEs (CCUEs) or cluster-edge UEs (CEUEs). Correspondingly, the total available spectrum band is divided as CCUE-band and CEUE-band, which are occupied by P-BSs associated with CCUEs and CEUEs, respectively. Unlike P-BSs, the F-BSs randomly access the CEUE-band and CCUE-band with a given access portion factor. By using the method from stochastic geometry and the distinguishing feature of mm-Wave, such as directional beamforming with sectored antenna model and modified line of sight ball model for blockage, the UE association probability and Laplace transforms of interference are derived. This yields the derivations of downlink transmission rates. The numerical results show that the proposed clustered heterogeneous scheme with the cluster-UE classification outperforms the traditional ones without the UE classification. This approach not only captures the UE-BS coupling and the non-homogeneous nature of multi-tier networks, but also restricts the severe intra-cluster interference due to clustered BSs.

Pilot contamination is the bottleneck of the massive MIMO systems. This paper proposed a pilot allocation scheme that combined user location and spatial distance. All users were clustered according to whether the angle of arrival overlapped, and classified the user sets whose arrival angles did not overlap with each other into n sets. Then the remaining users wh o did not satisfy the condition were classified into the n + 1 classes. It assigned n sets of pilot sequences to the first n types of user sets. For users in then+ 1 class set, the paper defined an interference function to construct a weighted graph. At the same time, it derived the user's expression of the uplink spectrum efficiency. The simulation results show that the scheme can significantly improve the spectrum efficiency. [ABSTRACT FROM AUTHOR]

Aiming at the problem of pilot contamination in massive MIMO system,a dynamic pilot assignment strategy based on user classification was proposed.The target cell user was divided into the central user and the edge user based on the variance of target cell user signal intensity.The central user used the maximum C/I algorithm to assign pilots to maximize the channel capacity of the central user.The edge user used greedy algorithm to assign pilots to improve the signal to interference and noise ratio of each edge user.Simulation results show that the proposed strategy can improve the channel capacity of the system while improving the signal to interference and noise ratio of the edge user.

In this paper, a novel covariance-based channel feedback mechanism is investigated for frequency division duplexing (FDD) massive multi-input multi-output (MIMO) systems. The concept capitalizes on the notion of user statistical separability which was hinted in several prior works in the massive antenna regime but has not fully exploited so far. We propose a hybrid statistical-instantaneous feedback mechanism where the users are separated into two classes of feedback design based on their channel covariance. Under the hybrid framework, each user either operates on a statistical feedback mode or quantized instantaneous channel feedback mode. The key challenge lies in the design of a covariance-aware classification algorithm which can handle the complex mutual interactions among all users. The classification is derived from rate bound principles and a precoding method is also devised under the mixed statistical and instantaneous feedback model. Simulations are performed to validate our analytical results and illustrate the sum rate advantages of the proposed feedback scheme under a global feedback overhead constraint. [ABSTRACT FROM AUTHOR]

The specificity of social services of the Internet communication is rapidly responding to events in society and the instant evaluation of the events of a group of users. This method of Internet communication affects the accuracy and truthfulness of the information increases the chances to use manipulation techniques of communication and a source of diverse information security threats. In this paper the analytics of social services of the internet communication and its user classification is proposed. The recommended users' classification helps media projects owners to create and manage these projects. [ABSTRACT FROM AUTHOR]

More and more network media users concern about their privacy issues since they know that their network behaviors are being observed, and thus the observable users’ data are not reliable and sufficient in this case. How to evaluate the true quality of experience (QoE) of the privacy-aware users has become a significant technical challenge because of the most majority of existing data-driven QoE evaluation schemes based on the premise of the true and adequate users’ observations. To get over this dilemma, this paper proposes a systematic and robust QoE evaluation scheme with unreliable and insufficient observation data. Specifically, we first translate the subjective privacy-aware QoE evaluation problem into an objective rational user analysis procedure. Then, a semantics-based similarity measurement for multidimensional correlation analysis is constructed to classify the observable data. Subsequently, the highlight of this paper lies in proposing a class-level joint user classification and data cleaning strategy by frequently updating the training processes. Through elaborately designing an iterative framework, it can effectively resolve the data sparsity and inconsistency problems due to the user privacy-aware preferences. Importantly, we also introduce an efficient QoE model construction method for online implementation, and numerical results validate its efficiency for different kinds of privacy-aware users. [ABSTRACT FROM AUTHOR]

Aiming to aid Electric-Power Industry to accurately understand users, hybrid learning model based ensemble learning algorithms for recognizing user to be sensitive to electric charge is proposed in this paper. On the basis of big data presented by CCF competition sponsor in China, with some excellent technology or algorithm such as JieBa, SFFS, etc., we extract many key features from data set and successfully draw a portrait for users who pay close attention to electric charge. Furthermore, machine learning algorithms and the strategy selection model related to them are investigated. The feasibility that hybrid learning model combining several ensemble learning algorithms can substantially improve classification accuracy are proved from theoretical level. Then the details of implementing hybrid learning model are described in the paper. Lastly, the hybrid learning model named Stacking is achieved, which yields better performance in contrast to the state-of-the-art competitors. The experimental results indicate that Stacking has both high precision and recall with 0.8 and 0.85 respectively. Furthermore the F1 score of Stacking evaluation is 0.823. [ABSTRACT FROM AUTHOR]

The vast amounts of mobile communication data collected by mobile operators can provide important insights regarding epidemic transmission or traffic patterns. By analyzing historical data and extracting user location information, various methods can be used to predict the mobility of mobile users. However, existing prediction algorithms are mainly based on the historical data of all users at an aggregated level and ignore the heterogeneity of individual behavior patterns. To improve prediction accuracy, this paper proposes a weighted Markov prediction model based on mobile user classification. The trajectory information of a user is extracted first by analyzing real mobile communication data, where the complexity of a user’s trajectory is measured using the mobile trajectory entropy. Second, classification criteria are proposed based on different user behavior patterns, and all users are classified with machine learning algorithms. Finally, according to the characteristics of each user classification, the step threshold and the weighting coefficients of the weighted Markov prediction model are optimized, and mobility prediction is performed for each user classification. Our results show that the optimized weighting coefficients can improve the performance of the weighted Markov prediction model.

Social networks are progressively being considered as an intense thought for learning. Particularly in the research area of Intelligent Tutoring Systems, they can create intuitive, versatile and customized e-learning systems which can advance the learning process by revealing the capacities and shortcomings of every learner and by customizing the correspondence by group profiling. In this paper, the primary idea is the affect recognition as an estimation of the group profiling process, given that the fact of knowing how individuals feel about specific points can be viewed as imperative for the improvement of the tutoring process. As a testbed for our research, we have built up a prototype system for recognizing the emotions of Facebook users. Users' emotions can be neutral, positive or negative. A feeling is frequently presented in unpretentious or complex ways in a status. On top of that, data assembled from Facebook regularly contain a considerable measure of noise. Indeed, the task of automatic affect recognition in online texts turns out to be more troublesome. Thus, a probabilistic approach of Rocchio classifier is utilized so that the learning process is assisted. Conclusively, the conducted experiments confirmed the usefulness of the described approach.