|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | Multi-Layer Graph Generative Model Using AutoEncoder for Recommendation Systems显示文摘Given the glut of information on the web,it is crucially important to have a system,which will parse the information appropriately and recommend users with relevant information,this class of systems is known as Recommendation Systems(RS)-it is one of the most extensively used systems on the web today.Recently,Deep Learning(DL)models are being used to generate recommendations,as it has shown state-of-the-art(SoTA)results in the field of Speech Recognition and Computer Vision in the last decade.However,the RS is a much harder problem,as the central variable in the recommendation system’s environment is the chaotic nature of the human’s purchasing/consuming behaviors and their interest.These user-item interactions cannot be fully represented in the Euclidean-Space,as it will trivialize the interaction and undermine the implicit interactions patterns.So to preserve the implicit as well as explicit interactions of user and items,we propose a new graph based recommendation framework.The fundamental idea behind this framework is not only to generate the recommendations in the unsupervised fashion but to learn the dynamics of the graph and predict the short and long term interest of the users.In this paper,we propose the first step,a heuristic multi-layer high-dimensional graph which preserves the implicit and explicit interactions between users and items using SoTA Deep Learning models such as AutoEncoders.To generate recommendation from this generated graph a new class of neural network architecture-Graph Neural Network-can be used. | Syed Falahuddin Quadri Xiaoyu Li Desheng Zheng Muhammad Umar Aftab Yiming Huang | 2019 | Journal on Big Data2019，1，1： | 1 |
| 2 | Analyzing and Enabling the Harmonious Coexistence of Heterogeneous Industrial Wireless Networks显示文摘Nowadays multiple wireless communication systems operate in industrial environments side by side.In such an environment performance of one wireless network can be degraded by the collocated hostile wireless network having higher transmission power or higher carrier sensing threshold.Unlike the previous research works which considered IEEE 802.15.4 for the Industrial Wireless communication systems(iWCS)this paper examines the coexistence of IEEE 802.11 based iWCS used for delay-stringent communication in process automation and gWLAN(general-purpose WLAN)used for non-real time communication.In this paper,we present a Markov chain-based performance model that described the transmission failure of iWCS due to geographical collision with gWLAN.The presented analytic model accurately determines throughput,packet transaction delay,and packet loss probability of iWCS when it is collocated with gWLAN.The results of the Markov model match more than 90%with our simulation results.Furthermore,we proposed an adaptive transmission power control technique for iWCS to overcome the potential interferences caused by the gWLAN transmissions.The simulation results show that the proposed technique significantly improves iWCS performance in terms of throughput,packet transaction,and cycle period reduction.Moreover,it enables the industrial network for the use of delay critical applications in the presence of gWLAN without affecting its performance. | Bilal Khan Danish Shehzad Numan Shafi Ga-Young Kim Muhammad Umar Aftab | 2022 | Computers, Materials & Continua2022，，10： | 0 |
| 3 | Genome editing in cotton:challenges and opportunities显示文摘Cotton has enormous economic potential providing high-quality protein,oil,and fibre.A large increase in cotton output is necessary due to the world’s changing climate and constantly expanding human population.In the past,conventional breeding techniques were used to introduce genes into superior cotton cultivars to increase production and to improve quality.The disadvantages of traditional breeding techniques are their time-consuming,reliance on genetic differences that are already present,and considerable backcrossing.To accomplish goals in a short amount of time,contemporary plant breeding techniques,in particular modern genome editing technologies(GETs),can be used.Numerous crop improvement initiatives have made use of GETs,such as zinc-finger nucleases,transcription-activator-like effector nucleases,clustered regularly interspaced palindromic repeats(CRISPR),and CRISPR-associated proteins systems(CRISPR/Cas)-based technologies.The CRISPR/Cas system has a lot of potential because it combines three qualities that other GETs lack:simplicity,competence,and adaptability.The CRISPR/Cas mechanism can be used to improve cotton tolerance to biotic and abiotic stresses,alter gene expression,and stack genes for critical features with little possibility of segregation.The transgene clean strategy improves CRISPR acceptability addressing regulatory issues associated with the genetically modified organisms(GMOs).The research opportunities for using the CRISPR/Cas system to address biotic and abiotic stresses,fibre quality,plant architecture and blooming,epigenetic changes,and gene stacking for commercially significant traits are highlighted in this article.Furthermore,challenges to use of CRISPR technology in cotton and its potential for the future are covered in detail. | KHAN Zulqurnain KHAN Sultan Habibullah AHMED Aftab IQBAL Muhammad Umar MUBARIK Muhammad Salman GHOURI Muhammad Zubair AHMAD Furqan YASEEN Saba ALI Zulfiqar KHAN Asif Ali AZHAR Muhammad Tehseen | 2023 | Journal of Cotton Research2023，6，1： | 0 |
| 4 | A Compromise Programming to Task Assignment Problem in Software Development Project显示文摘The scheduling process that aims to assign tasks to members is a difficult job in project management.It plays a prerequisite role in determining the project’s quality and sometimes winning the bidding process.This study aims to propose an approach based on multi-objective combinatorial optimization to do this automatically.The generated schedule directs the project to be completed with the shortest critical path,at the minimum cost,while maintaining its quality.There are several real-world business constraints related to human resources,the similarity of the tasks added to the optimization model,and the literature’s traditional rules.To support the decision-maker to evaluate different decision strategies,we use compromise programming to transform multiobjective optimization(MOP)into a single-objective problem.We designed a genetic algorithm scheme to solve the transformed problem.The proposed method allows the incorporation of the model as a navigator for search agents in the optimal solution search process by transferring the objective function to the agents’fitness function.The optimizer can effectively find compromise solutions even if the user may or may not assign a priority to particular objectives.These are achieved through a combination of nonpreference and preference approaches.The experimental results show that the proposed method worked well on the tested dataset. | Ngo Tung Son Jafreezal Jaafar Izzatdin Abdul Aziz Bui Ngoc Anh Hoang Duc Binh Muhammad Umar Aftab | 2021 | Computers, Materials & Continua2021，，12： | 0 |
| 5 | Multi Sensor-Based Implicit User Identification显示文摘Smartphones have ubiquitously integrated into our home and work environments,however,users normally rely on explicit but inefficient identification processes in a controlled environment.Therefore,when a device is stolen,a thief can have access to the owner’s personal information and services against the stored passwords.As a result of this potential scenario,this work proposes an automatic legitimate user identification system based on gait biometrics extracted from user walking patterns captured by smartphone sensors.A set of preprocessing schemes are applied to calibrate noisy and invalid samples and augment the gait-induced time and frequency domain features,then further optimized using a non-linear unsupervised feature selection method.The selected features create an underlying gait biometric representation able to discriminate among individuals and identify them uniquely.Different classifiers are adopted to achieve accurate legitimate user identification.Extensive experiments on a group of 16 individuals in an indoor environment show the effectiveness of the proposed solution:with 5 to 70 samples per window,KNN and bagging classifiers achieve 87–99%accuracy,82–98%for ELM,and 81–94%for SVM.The proposed pipeline achieves a 100%true positive and 0%false-negative rate for almost all classifiers. | Muhammad Ahmad Rana Aamir Raza Manuel Mazzara Salvatore Distefano Ali Kashif Bashir Adil Khan Muhammad Shahzad Sarfraz Muhammad Umar Aftab | 2021 | Computers, Materials & Continua2021，，8： | 0 |
您所在的机构暂未开通此部分的功能，请联系管理员