- Game Theory and Applications
- Opinion Dynamics and Social Influence
- Complex Network Analysis Techniques
- Complexity and Algorithms in Graphs
- Cloud Computing and Resource Management
- Optimization and Search Problems
- Distributed and Parallel Computing Systems
- Game Theory and Voting Systems
- Peer-to-Peer Network Technologies
- Cryptography and Data Security
- Mobile Crowdsensing and Crowdsourcing
- Data Management and Algorithms
- Computational Geometry and Mesh Generation
- Stock Market Forecasting Methods
- IoT and Edge/Fog Computing
- Topic Modeling
- Anomaly Detection Techniques and Applications
- Caching and Content Delivery
- Distributed systems and fault tolerance
- Sports Analytics and Performance
- Digital Platforms and Economics
- COVID-19 diagnosis using AI
- Auction Theory and Applications
- Expert finding and Q&A systems
- Machine Learning and Data Classification
Sharif University of Technology
2014-2025
Universiti Teknologi MARA
2024
In today's digital age, Convolutional Neural Networks (CNNs), a subset of Deep Learning (DL), are widely used for various computer vision tasks such as image classification, object detection, and segmentation. There numerous types CNNs designed to meet specific needs requirements, including 1D, 2D, 3D CNNs, well dilated, grouped, attention, depthwise convolutions, NAS, among others. Each type CNN has its unique structure characteristics, making it suitable tasks. It's crucial gain thorough...
We introduce a network creation game in which each player (vertex) has fixed budget to establish links other players. In this model, link unit price, and agent tries minimize its cost, is either eccentricity or total distance players the underlying (undirected) graph of created network. Two versions are studied: MAX version, cost incurred vertex maximum between vertices, and, SUM sum distances vertices. prove that both pure Nash equilibria exist, but problem finding best response NP-hard....
<title>Abstract</title> This study formalizes a computational model to simulate classical Persian poets' dynamics of influence through constructing multi-dimensional similarity network. Using rigorously curated dataset based on Ganjoor's corpus, we draw upon semantic, lexical, stylistic, thematic, and metrical features demarcate each poet's corpus. Each is contained within weighted matrices, which are then appended generate an aggregate graph showing poet-to-poet influence. Further network...
We consider a network creation game in which, each player (vertex) has limited budget to establish links other players. In our model, link unit cost and agent tries minimize its which is local diameter or total distance players the (undirected) underlying graph of created network. Two variants are studied: MAX version, incurred vertex maximum between that vertices, SUM sum distances vertices. prove both versions pure Nash equilibria exist, but problem finding best response NP-hard.
Community Question-Answering websites, such as StackOverflow and Quora, expect users to follow specific guidelines in order maintain content quality. These systems mainly rely on community reports for assessing contents, which has serious problems the slow handling of violations, loss normal experienced users' time, low quality some reports, discouraging feedback new users. Therefore, with overall goal providing solutions automating moderation actions Q&A we aim provide a model predict 20 or...
Social norms are a core concept in social sciences and play critical role regulating society's behavior. Organizations even governmental bodies use this component to tackle varying challenges the society, as it is less costly alternative establishing new laws regulations. networks an important effective infrastructure which can evolve. Therefore, there need for theoretical models studying spread of networks. In paper, by using intrinsic properties norms, we redefine tune Rescorla-Wagner...
With the widespread access of people to Internet and increasing usage social networks in all nations, have become a new source study cultural similarities differences. We identified major issues traditional methods data collection cross-cultural studies: difficulty from many limited number samples, negative effects translation, positive self-enhancement illusion, few unreported problems. These are either causing perform or impacts on validity final results. In this paper, we propose...
In today's digital age, Convolutional Neural Networks (CNNs), a subset of Deep Learning (DL), are widely used for various computer vision tasks such as image classification, object detection, and segmentation. There numerous types CNNs designed to meet specific needs requirements, including 1D, 2D, 3D CNNs, well dilated, grouped, attention, depthwise convolutions, NAS, among others. Each type CNN has its unique structure characteristics, making it suitable tasks. It's crucial gain thorough...
Controlling networked dynamical systems is a challenging endeavor, specifically keeping in mind the fact that many scenarios actors engaged dynamism behave selfishly, only taking into account their own individual utility. This setting has been widely studied field of game theory. One way we can control system dynamics through use parameters are at our disposal, but finding optimal values for these complex and time consuming. In this paper relation between network structural properties to...
Views Icon Article contents Figures & tables Video Audio Supplementary Data Peer Review Share Twitter Facebook Reddit LinkedIn Tools Reprints and Permissions Cite Search Site Citation A. Z. Bahtar, S. Daud, Yazid, M. Fazli, G. Muthusamy, R. Kamis, H. Harun, Awang; The influence of e-service quality towards post-purchase intention on shopee. AIP Conf. Proc. 19 April 2024; 2799 (1): 020018. https://doi.org/10.1063/5.0183800 Download citation file: Ris (Zotero) Reference Manager EasyBib...
Financial markets are integral to a country's economic success, yet their complex nature raises challenging issues for predicting behaviors. There is growing demand an integrated system that explores the vast and diverse data in financial reports with powerful machine-learning models analyze suggest appropriate investment strategies. This research provides end-to-end decision support (DSS) pervasively covers stages of gathering, cleaning, modeling stock's fundamental alongside macroeconomic...
In this paper we study envy free pricing problem in general graphs where there is not a seller every graph's nodes. We assume unique establishment cost for initiating store each node and wish to find an optimal set of nodes which would make the maximum profit by stores them. Our model motivated from observation that same product has different prices locations also establishing any store. consider both these issues our model: first should establish stores, second at what price sell items them...