User Clustering Research Articles

Power minimization for GSIC-based uplink cell-free massive MIMO-NOMA systems

Non-orthogonal multiple access (NOMA) and multiple-input multiple-output (MIMO) are considered promising techniques to satisfy the demands for high spectrum efficiency and massive connectivity in future wireless communication. In this paper, a novel framework to realize transmission in cell-free massive MIMO-NOMA system with deep integration between MIMO and NOMA is proposed. A new method is developed to divide users into different groups according to their equivalent path loss, and then “group-level successive interference cancellation” (GSIC) is used to cancel the inter-group interference when demodulating the users. Based on the new framework, a rigorous closed-form expression of the achievable sum rate in uplink cell-free massive MIMO-NOMA system is derived. In addition, a parallel iterative method is used to obtain the best power control scheme. Simulations show that the proposed scheme can effectively reduce the total power consumption and outperforms the orthogonal multiple access (OMA) transmission and traditional SIC-NOMA schemes based on user clustering.

Machine Learning Empowered Resource Allocation in IRS Aided MISO-NOMA Networks

A novel framework of intelligent reflecting surface (IRS)-aided multiple-input single-output (MISO) non-orthogonal multiple access (NOMA) network is proposed, where a base station (BS) serves multiple clusters with unfixed number of users in each cluster. The goal is to maximize the sum rate of all users by jointly optimizing the passive beamforming vector at the IRS, decoding order, power allocation coefficient vector and number of clusters, subject to the rate requirements of users. In order to tackle the formulated problem, a three-step approach is proposed. More particularly, a long short-term memory (LSTM) based algorithm is first adopted for predicting the mobility of users. Secondly, a K-means based Gaussian mixture model (K-GMM) algorithm is proposed for user clustering. Thirdly, a deep Q-network (DQN) based algorithm is invoked for jointly determining the phase shift matrix and power allocation policy. Simulation results are provided for demonstrating that the proposed algorithm outperforms the benchmarks, while the throughput gain of 35% can be achieved by invoking NOMA technique instead of orthogonal multiple access (OMA).

Election Fraud and Misinformation on Twitter: Author, Cluster, and Message Antecedents

This study determined the antecedents of diffusion scope (total audience), speed (number of adopters/time), and shape (broadcast vs. person-to-person transmission) for true vs. fake news about a falsely claimed stolen 2020 US Presidential election across clusters of users that responded to one another’s tweets (“user clusters”). We examined 31,128 tweets with links to fake vs. true news by 20,179 users to identify 1,069 user clusters via clustering analysis. We tested whether attributes of authors (experience, followers, following, total tweets), time (date), or tweets (link to fake [vs. true] news, retweets) affected diffusion scope, speed, or shape, across user clusters via multilevel diffusion analysis. These tweets showed no overall diffusion pattern; instead, specific explanatory variables determined their scope, speed, and shape. Compared to true news tweets, fake news tweets started earlier and showed greater broadcast influence (greater diffusion speed), scope, and person-to-person influence. Authors with more experience and smaller user clusters both showed greater speed but less scope and less person-to-person influence. Likewise, later tweets showed slightly more broadcast influence, less scope, and more person-to-person influence. By contrast, users with more followers showed less broadcast influence but greater scope and slightly more person-to-person influence. These results highlight the earlier instances of fake news and the greater diffusion speed of fake news in smaller user clusters and by users with fewer followers, so they suggest that monitors can detect fake news earlier by focusing on earlier tweets, smaller user clusters, and users with fewer followers.

Multi-UAV Content Caching Strategy and Cooperative, Complementary Content Transmission Based on Coalition Formation Game.

The transmission of a large amount of video and picture content brings more challenges to wireless communication networks. Unmanned aerial vehicle (UAV)-aided small cells with active content caching deployed on cellular networks are recognized as a promising way to alleviate wireless backhaul and support flexible coverage. However, a UAV cannot operate for a long time due to limited battery life, and its caching capacity is also limited. For this, a multi-UAV content-caching strategy and cooperative, complementary content transmission among UAVs are jointly studied in this paper. Firstly, a user-clustering-based caching strategy is designed, where user clustering is based on user similarity, concurrently taking into consideration similarities in content preference and location. Then, cooperative, complementary content transmission between multiple UAVs is modeled as a coalition formation game (CFG) to maximize the utility of the whole network. Finally, the effectiveness of the proposed algorithms is demonstrated through numerical simulations.

Density-based user clustering in downlink NOMA systems

Density-based user clustering in downlink NOMA systems

RIS-Assisted Joint Transmission in a Two-Cell Downlink NOMA Cellular System

This paper investigates the integration of reconfigurable intelligent surface (RIS) with downlink non-orthogonal-multiple-access (NOMA) in a multi-user two-cell network assisted by the joint-transmission coordinated multipoint (JT-CoMP). Specifically, the RIS is deployed at the edge of two adjacent cells to assist the JT-CoMP from these two cells to multiple far NOMA users located at their edges. Under this setup, we jointly optimize the power allocation (PA) coefficients at the base stations (BSs), the user clustering (UC) policy, and the phase-shift (PS) matrix of the RIS with the objective of maximizing the network sum-rate subject to a target quality-of-service, defined in terms of the minimum required data rate at each cellular user, and the successive interference cancellation (SIC) constraints. The formulated problem ends to be a non-convex mixed-integer non-linear program that is difficult to be solved in a straightforward manner. To alleviate this issue, and with the aid of alternating optimization (AO), the original optimization problem is decomposed into two sub-problems, a joint PA and UC sub-problem and a PS sub-problem, that are solved in an alternating way. For the first sub-problem, we invoke the bi-level optimization approach to decouple the PA sub-problem from the UC sub-problem. For the PA sub-problem, closed-form expressions for the optimal PA coefficients are derived. On the other hand, the UC problem is projected to multiple 2-dimensional assignment problems, each of which is solved using the Hungarian method. Finally, the PS sub-problem is formulated as a difference-of-convex problem and an efficient solution is obtained using the successive convex approximation technique. The numerical results reveal that the network sum-rate of the proposed RIS-assisted CoMP NOMA networks outperforms the conventional CoMP NOMA scheme without the assistance of the RIS, the RIS-assisted CoMP orthogonal multiple access (OMA) scheme, and RIS-assisted NOMA scheme, especially for low transmit power from the BSs.

Filter bubble effect in the multistate voter model.

Social media influence online activity by recommending to users content strongly correlated with what they have preferred in the past. In this way, they constrain users within filter bubbles strongly limiting their exposure to new or alternative content. We investigate this type of dynamics by considering a multistate voter model where, with a given probability λ, a user interacts with "personalized information," suggesting the opinion most frequently held in the past. By means of theoretical arguments and numerical simulations, we show the existence of a nontrivial transition between a region (for small λ) where a consensus is reached and a region (above a threshold λc) where the system gets polarized and clusters of users with different opinions persist indefinitely. The threshold always vanishes for large system size N, showing that a consensus becomes impossible for a large number of users. This finding opens new questions about the side effects of the widespread use of personalized recommendation algorithms.

Dynamic User Clustering and Optimal Power Allocation in UAV-Assisted Full-Duplex Hybrid NOMA System

This paper investigates unmanned aerial vehicles (UAVs)-assisted full-duplex (FD) non-orthogonal multiple access (NOMA) system based cellular network, aiming to improve overall sum-rate throughput of the system through dynamic user clustering, optimal UAV placement and power allocation. Since each UAV operates in FD mode, self-interference (SI), co-channel interference (CCI), inter-UAV interference (IUI) and intra-node interference (INI) dominate the system&#x2019;s performance. Consequently, we propose an unconventional two-stage dynamic user clustering for user nodes (UNs) to reduce the cross-interference in multi-UAV aided FD-NOMA system. Particularly, all UNs are initially clustered into <inline-formula> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> clusters using k-means clustering in the first stage where each cluster is served by an UAV. Furthermore, each cluster is further divided into sub-clusters and each sub-clusters are operated in FD-NOMA scheme. Finally, to control interferences, a sum-rate throughput maximization problem is formulated for each UAV to jointly optimize uplink and downlink power allocation and UAV placement. The joint optimization problem is non-convex and difficult to solve directly, for which we decoupled the original problem by addressing UAV placement and power allocation separately. We first fix the UAV position and then solve the problem iteratively using successive convex approximation (SCA) method. By utilizing brute-force search algorithm, an optimal UAV placement is later performed which corresponds to maximum possible sum-rate throughput. Simulation results demonstrate that the proposed solution for the considered FD-NOMA system outperforms the conventional schemes.

Joint optimization scheme based on beam selection and interference cancellation in lens millimeter wave NOMA systems

Non-orthogonal multiple access (NOMA) has been integrated in millimeter-wave (mmWave) Massive MIMO systems to further enhance the spectrum efficiency, but NOMA has not been fully considered in lens mmWave systems. The fusion of these two technologies requires the joint design of beam selection and interference cancellation. In addition, when users follow the spatial cluster distribution, although the user clustering schemes based on K-means algorithm have been applied, the influence of the virtual and actual cluster center users on achievable sum rate has not been differentiated and analyzed in detail. To solve the above problems, a joint optimization scheme is proposed to maximize the achievable sum rate. The optimization problem is NP-hard, which is solved by using the divide-and-conquer approach. Concretely, based on the signal power loss analysis of directional deviation, a beam selection algorithm is designed for inter-cluster interference cancellation based on the K-means algorithm. Further for intra-cluster interference cancellation, a power allocation algorithm based on the bisection method is designed to guarantee the fairness of users in each cluster. The simulation results show that the proposed scheme offers a significant performance improvement in terms of both achievable sum rate and energy efficiency, and it is suitable for the large-scale user scenario due to its low complexity.

Clustering Merchants and Accurate Marketing of Products Using the Segmentation Tree Vector Space Model

Using social commerce users as the data source, a reasonable and effective interest expression mechanism is used to construct an interest graph of sample users to achieve the purpose of clustering merchants and users as well as realizing accurate marketing of products. By introducing an improved vector space model, the segmentation tree vector space model, to express the interests of the target user group and, on this basis, using the complex network analysis tool Gephi to construct an interest graph, based on the user interest graph, we use Python to implement the K-means algorithm and the users of the sample set according to interest topics for community discovery. The experimental results show that the interests of the sample users are carefully divided, each user is divided into different thematic communities according to different interests, and the constructed interest graph is more satisfactory. The research design of the social commerce user interest mapping scheme is highly feasible, reasonable, and effective and provides new ideas for the research of interest graph, and the boundaries of thematic communities based on interests are clear.

Design of Garment Style Recommendation System Based on Interactive Genetic Algorithm.

Recommender systems provide users with product information and suggestions, which has gradually become an important research tool in e-commerce IT technology, which has attracted a lot of attention of researchers. Collaborative filtering recommendation technology has been the most successful recommendation technology so far, but there are two major problems—recommendation quality and scalability. At present, research at home and abroad mainly focuses on recommendation quality, and there is less discussion on scalability. The scalability problem is that as the size of the system increases, the response time of the system increases to a point where users cannot afford it. Existing solutions often result in a significant drop in recommendation quality while reducing recommendation response time. In this paper, the clustering analysis subsystem based on the genetic algorithm is innovatively introduced into the traditional collaborative filtering recommendation system, and its design and implementation are given. In addition, when obtaining the nearest neighbors, only the clustered users of the target user are searched, making it a collaborative filtering recommender system based on genetic clustering. The experimental results show that the response time of the traditional collaborative filtering recommender system increases linearly with the increase in the number of users while the response time of the collaborative filtering recommender system based on genetic clustering remains unchanged with the increase in the number of users. On the other hand, the recommendation quality of the collaborative filtering recommender system based on genetic clustering is basically not degraded compared with that of the traditional collaborative filtering recommender system. Therefore, the collaborative filtering recommender system based on genetic clustering can effectively solve the scalability problem of the collaborative filtering recommender system.

GHRS: Graph-based hybrid recommendation system with application to movie recommendation

Research about recommender systems emerges over the last decade and comprises valuable services to increase different companies’ revenue. While most existing recommender systems rely either on a content-based approach or a collaborative approach, there are hybrid approaches that can improve recommendation accuracy using a combination of both approaches. Even though many algorithms are proposed using such methods, it is still necessary for further improvement. This paper proposes a recommender system method using a graph-based model associated with the similarity of users’ ratings in combination with users’ demographic and location information. By utilizing the advantages of Autoencoder feature extraction, we extract new features based on all combined attributes. Using the new set of features for clustering users, our proposed approach (GHRS) outperformed many existing recommendation algorithms on recommendation accuracy. Also, the method achieved significant results in the cold-start problem. All experiments have been performed on the MovieLens dataset due to the existence of users’ side information.

UAV-Assisted Cooperative &amp; Cognitive NOMA: Deployment, Clustering, and Resource Allocation

Cooperative and cognitive non-orthogonal multiple access (CCR-NOMA) has been recognized as a promising technique to overcome issues of spectrum scarcity and support massive connectivity envisioned in next-generation wireless networks. In this paper, we investigate the deployment of an unmanned aerial vehicle (UAV) as a relay that fairly serves a large number of secondary users in a hot-spot region. The UAV deployment algorithm must jointly account for user clustering, channel assignment, and resource allocation sub-problems. We propose a solution methodology that obtains user clustering and channel assignment based on the optimal resource allocations for a given UAV location. To this end, we derive closed-form optimal power and time allocations and show it delivers optimal max-min fair throughput by consuming less energy and time than geometric programming. Based on optimal resource allocation, the optimal coverage probability is also provided in closed-form, which takes channel estimation errors, hardware impairments, and primary network interference into account. The optimal coverage probabilities are used by the proposed max-min fair user clustering and channel assignment approaches. The results show that the proposed method achieves 100% accuracy in more than five orders of magnitude less time than the optimal benchmark.

A Gradual Resource Allocation Technique for Massive MIMO-NOMA

In this letter, a new perspective for channel allocation and user pairing is proposed for a crowded downlink system using massive multiple-input–multiple-output, nonorthogonal multiple access (NOMA), and zero-forcing beamforming. The proposed user clustering technique allocates users to subbands and beams by a gradual relaxation of the correlation constraints in such a way to enhance the conditioning of the channel matrix. Channel measurements are performed in this context to validate the proposed technique in an experimental setup. Results show an important gain in performance in terms of both throughput and fairness with respect to orthogonal multiple access and to two state-of-the-art NOMA clustering methods.

Synchronized Action Framework for Detection of Coordination on Social Media

The study of the coordinated manipulation of conversations on social media has become more prevalent as social media’s role in amplifying misinformation, hate, and polarization has come under greater scrutiny. We discuss how successful generalized coordination detection algorithms could be used to reinforce existing power imbalances, such as those between marginalized groups and government agencies. We propose an alternative method of identifying manipulation—detecting synchronized actions—which reduces this risk. We further consider how responsible coordination detection may be carried out by analyzing synchronized actions. We propose a synchronized action framework for detecting automated coordination by constructing and analyzing multi-view networks. We validate our framework by examining a large Twitter dataset surrounding the Reopen America conversation from 2020. We first discover three simple coordinated campaigns, and then investigate synchronized actions between users discussing the protests that could be consistent with covert coordination. This task is far more complex than examples evaluated in prior work, which demonstrates the need for our multi-view approach. Next, we identify a cluster of suspicious users and detail the activity of three members. These three users amplify protest messages using the same hashtags at very similar times, though they all focus on different states. This analysis highlights the potential usefulness of coordination detection algorithms in investigating amplification, as well as the need to carefully and responsibly deploy such tools.

Mobile Learning Platform in Cloud Computing with Information Security and Android System

In order to meet the real-time communication between teachers and students and improve students' interest and efficiency in learning, this paper designs a mobile learning communication system based on information security and Android system with cloud computing. By using the method of combining database records and web logs to mine the user's browsing records and behaviors, these implicit user behaviors are transformed into explicit user evaluations of the project. Then, the cosine similarity calculation method is used to calculate users and the similarity between the users. The users are clustered by the K-means clustering method, so that the users are automatically divided into several user clusters according to their behaviors. Finally, the user's nearest neighbor score is used to predict the pair. Based on the above method, a mobile learning communication system based on Android is realized, and the system basically meets the functional needs of users. The development of this system not only promotes the mutual communication between students but also facilitates the students' learning, which has a certain promoting effect on the improvement of their academic performance.

Machine Learning Based Content-Agnostic Viewport Prediction for 360-Degree Video

Accurate and fast estimations or predictions of the (near) future location of the users of head-mounted devices within the virtual omnidirectional environment open a plethora of opportunities in application domains such as interactive immersive gaming and tele-surgery. Therefore, the past years have seen growing attention to models for viewport prediction in 360֯ environments. Among the approaches, content-agnostic, trajectory-based methods have the potential to provide very fast solutions, as they do not require complex analysis of the videos to provide a prediction. However, accurate trajectory-based viewport prediction is rather difficult due to the intrinsic variability in user behaviour. Furthermore, even when making use of machine learning, current approaches tend to be brute-force and heavily tailored to specific datasets with little comparison to existing benchmarks or publicly available studies. This article presents a generic, content-agnostic viewport prediction method consisting of a window-based approach combined with a preprocessing system to classify behavioural patterns in terms of user clustering and trajectory correlation. Moreover, as the state of the art does not provide a comparative analysis of different approaches, this work contributes to this. Based on the obtained results, a combined prediction model is proposed and evaluated. Our method shows a 36.8% to 53.9% improvement when compared to the static prediction baseline for a prediction horizon of 8 seconds. In addition, a 11.5% to 24.0% improvement to a brute-force machine learning prediction approach is obtained. As such, this work contributes towards the creation of more generic and structured solutions for content-agnostic viewport prediction in terms of data representation, preprocessing and modelling.

Joint Beam-Forming, User Clustering and Power Allocation for MIMO-NOMA Systems.

In this paper, we consider the optimal resource allocation problem for multiple-input multiple-output non-orthogonal multiple access (MIMO-NOMA) systems, which consists of beam-forming, user clustering and power allocation, respectively. Users can be divided into different clusters, and the users in the same cluster are served by the same beam vector. Inter-cluster orthogonality can be guaranteed based on multi-user detection (MUD). In this paper, we propose a three-step framework to solve the multi-dimensional resource allocation problem. In step 1, we propose a beam-forming algorithm for a given user cluster. Specifically, fractional transmitting power control (FTPC) is applied for intra-cluster power allocation. The considered beam-forming problem can be transformed into a non-constrained one and the limited-memory Broyden–Fletcher–Goldfarb–Shanno (L-BFGS) method is applied to obtain the optimal solution. In step 2, optimal user clustering is further considered. Channel differences and correlations are both involved in the design of user clustering. By assigning different weights to the two factors, we can produce multiple candidate clustering schemes. Based on the proposed beam-forming algorithm, beam-forming can be done for each candidate clustering scheme to compare their performances. Moreover, based on the optimal user clustering and beam-forming schemes, in step 3, power allocation can be further optimized. Specifically, it can be formalized as a difference of convex (DC) programming problem, which is solved by successive convex approximation (SCA) with strong robustness. Simulations results show that the proposed scheme can effectively improve spectral efficiency (SE) and edge users’ data rates.

E-commerce Recommender System Using PCA and K-Means Clustering

Recently, recommender system has an important role in e-commerce to market products for users. One of recommender system approach that used in e-commerce is Collaborative Filtering. This system works by providing product recommendations based on products liked by other users who have similar preferences. However, sparse conditions in user data will cause sparsity problems, namely the system is difficult to provide recommendations because of the lack of important information needed. Therefore, we propose an e-commerce product recommendation system based on Collaborative Filtering using Principal Component Analysis (PCA) and K-Means Clustering. K-Means is used to overcome sparsity problems and to form user clusters to reduce the amount of data that needs to be processed. While PCA is used to reduce data dimensions and improve clustering performance of K-Means. The test results using the sports product dataset on the Olist e-commerce show that the proposed system has a lower RMSE value compared to other methods. For the number of neighbors of 10, 20, 30, and 40, our system obtains values of 0.771806, 0.75747, 0.75304, 0.75304, and 0.75270.&#x0D;

Efficient Algebraic Multigrid Methods for Multilevel Overlapping Coclustering of User-Item Relationships

Various digital data sets that encode user-item relationships contain a multilevel overlapping cluster structure. The user-item relation can be encoded in a weighted bipartite graph and uncovering these overlapping coclusters of users and items at multiple levels in the bipartite graph can play an important role in analyzing user-item data in many applications. For example, for effective online marketing, such as placing online ads or deploying smart online marketing strategies, identifying co-occurring clusters of users and items can lead to accurately targeted advertisements and better marketing outcomes. In this paper, we propose fast algorithms inspired by algebraic multigrid methods for finding multilevel overlapping cocluster structures of feature matrices that encode user-item relations. Starting from the weighted bipartite graph structure of the feature matrix, the algorithms use agglomeration procedures to recursively coarsen the bipartite graphs that represent the relations between the coclusters on increasingly coarser levels. New fast coarsening routines are described that circumvent the bottleneck of all-to-all similarity computations by exploiting measures of direct connection strength between row and column variables in the feature matrix. Providing accurate coclusters at multiple levels in a manner that can scale to large data sets is a challenging task. In this paper, we propose heuristic algorithms that approximately and recursively minimize normalized cuts to obtain coclusters in the aggregated bipartite graphs on multiple levels of resolution. Whereas the main novelty and focus of the paper lies in algorithmic aspects of reducing computational complexity to obtain scalable methods specifically for large rectangular user-item matrices, the algorithmic variants also define several new models for determining multilevel coclusters that we justify intuitively by relating them to principles that underlie collaborative filtering methods for user-item relationships. Experimental results show that the proposed algorithms successfully uncover the multilevel overlapping cluster structure for artificial and real data sets. Summary of Contribution: This paper develops new and efficient computational methods for finding the multilevel overlapping cocluster structure of feature matrices that encode user-item relationships. We base our approach on the use of pairwise similarity measures between features, seeking clusters of points that are similar to each other and dissimilar from the points outside the cluster. We approximately solve the problem of finding optimal overlapping coclusters on multiple levels by employing a framework that is based on efficient multilevel methods that have been used previously to solve sparse linear systems and to cluster graphs. Our main contribution is that we extend these methods in efficient manners to find coclusters in the bipartite graphs that encode common and important user-item relationships or social network relations. The novel methods that we propose are inherently scalable to large problem sizes and are naturally able to uncover overlapping coclusters at multiple levels, whereas existing methods generally only find coclusters at the fine level. We illustrate the algorithm and its performance on some standard test problems from the literature and on a proof-of-concept real-world data set that relates LinkedIn users to their skills and expertise.