Sharif Digital Repository

The competition for staying in the business world has intensified today with the rise of open-source and commercial software. As long as a software is tailor-made to suit the needs of users, it is so-called alive and can stay in the competition. So the maintenance phase is necessary to make changes to the software to meet the needs of users. To reduce costs associated with this phase, it is necessary to avoid software bugs. One way to avoid software bugs is to use peer code review. Peer code review has been recognized as one of the best software engineering principles of the last 35 years. This principle helps maintain the quality of the code due to changes made to parts of the code that... 

Representation learning methods on graphs have enabled using machine learning methods on graphs' discrete structure by transferring them to a continuous domain. As graphs' structures are not always static and may evolve through time, dynamic representation learning methods have recently gained scholars' attention. Several methods have been proposed to enable the model to update the embeddings graph changes, or new interactions happen between nodes. These online methods could significantly reduce the learning time by refreshing the model as the changes occur, so we don't need to retrain the model with the complete graph information. Moreover, by using the temporal information of interactions,... 

Stock price patterns are a technical analysis approach to forecast future trends with tremendous practical benefits. However, the current algorithms solely rely on machine learning techniques and deep neural networks which could be a problem in countries where data sets such as these are not available. We propose an algorithm based on geometry and mathematics for this problem, leading to an O(n^3logn + n^2k) complexity, where k is the number of triangular patterns  

Video games with respect to an ever Increasing player pool and Industry growth, have attracted a lot of attention in recent years. Dota 2, as one of the most successful games both in casual gamers’ community and E-sport community, is considered as a proper case study, however, most of the research done was limited to predicting games’ outcome. Despite The popularity, the rather unintelligent AI of the game has made quite a frustrating experience for new players. In this research, with a novel approach, hero features are used to predict their future positions. For this purpose, 35 professional games are collected and analyzed and 601 features are extracted. Then, suitable features are... 

Deep learning algorithms combined with supervision rely heavily on labeled data, posing challenges in the data labeling process. Addressing this issue, researchers in the field of machine learning have focused on developing approaches to reduce the dependency on labeled data and improve the efficiency of data collection for labeling purposes. This thesis investigates the training of a classification model using data collected through a human-in-the-loop system. Notably, this research pioneers the application of active learning techniques to differentiate between political and non-political Persian tweets. The dataset introduced in this study is the sole available collection for this specific... 

Content delivery networks play a significant role in improving the quality of internet services by placing the necessary content closer to users on servers. Currently, over half of internet traffic is delivered through these networks to end users. The efficiency of a content delivery network depends on various parameters, including the type of requested content, workload distribution methods, network topology, routing algorithms, caching policies, network server configurations, and resource allocation (shared or dedicated hardware resources). Additionally, the requests made to a content delivery network vary based on the type of service and even the time of day, making optimization a... 

Software-defined networking (SDN) is a network management approach that allows centralized control of the network independently from network hardware. As a network management method, SDN provides many capabilities not traditionally found in existing hardware-based networks. However, one of the significant challenges of SDN is the placement of the controller within the network. In SDN, the central controller must be able to efficiently route all data flows separately to end devices. Therefore, the placement of the controller in the network is crucial. However, controller placement in the network poses a major challenge as it needs to be appropriately and optimally positioned to enhance... 

Over the past few years, deep reinforcement learning(DRL) has been given a lot of attention in finance for portfolio management. With the help of experts’ signals and historical price data, we have developed a new reinforcement learning(RL) method. The use of experts’ signals in tandem with DRL has been used before in finance, but we believe this is the first time this method has been used to solve the financial portfolio management problem. As our agent, we used the Proximal Policy Optimization(PPO) algorithm to process the reward and take actions in the environment. Our framework comprises a convolutional network to aggregate signals, a convolutional network for historical price data, and... 

In the world of software development, Code Review is one of the most vital processes to ensure code quality and security. The textual content features in code review comments play a significant role in assessing quality and guiding the review process. This research aims to examine the importance and role of these features in identifying anti-social comments and improving code review processes. In this study, we first challenge the concept of toxicity in code review comments, which had previously been accepted as a concept in the field of code review. We focus on enhancing and automating code review processes by accurately and reliably detecting anti-social comments based on relevant... 

Based on fog computing paradigm, new applications have become feasible through the use of hardware capabilities of smart phones. Many of these applications require a vast amount of computing and real-time execution should be guaranteed. Based on fog computing, in order to solve these problems in is necessary to offload heavy computing to servers with adequate hardware capabilities. On the other side, the offloading process causes time overhead and endangers the real-timeliness of the application. Also, because of the limited battery capacity of the handheld devices, energy consumption is very important and should be minimized.The usual proposed solution for this problem is to refactor the... 

Investors highly value the ability to predict the behavior of the capital market. Over time, various methods have been introduced to forecast the future of this market and anticipate its movements. A novel approach to achieving this is by developing data-driven decision support systems that can assist investors in making informed trading decisions. The opinions of experts play a crucial role in shaping people's perception of the market, which ultimately affects its final behavior. In this study, we have created a decision support system that can help investors by considering the complexities and meaningful relationships between different aspects of the problem. We have developed frameworks... 

Portfolio optimization is a crucial concept in financial engineering, focusing on the efficient management of investment portfolios. In the realm of financial markets, a portfolio refers to a collection of investments held by individuals or companies, encompassing diverse assets. Specifically, a stock portfolio consists solely of stocks. The primary objective of portfolio optimization methods is to maximize returns while controlling risks. Within Tehran’s Stock Market, valuable data pertaining to the stock portfolios of big shareholders and their historical changes can be obtained. This dataset contains vital information that can be leveraged to optimize portfolios over time and formulate... 

This paper is concerned with the Vehicle Routing Problem (VRP) in a fleet with customer loyalty. In this work, we consider priority for customers according to their loyalty and try to declare an objective function which concerns customers' satisfaction along other normal VRP objectives like route optimization  

The need for a fast, light and affordable payment method, was a reason to produce multiple payment channel network on various of networks. To deploy and expand these networks, we need to deploy a profit model for new customers on the betweenness they add to network. In this research we maximum profit solutions for network with fixed payment cost and network with reduced betweenness profit for longer distances.At first, we suggest for a limited fund, we decide to make channels with most betweenness profit and prove it's a np-hard problem. Then for limited charge of making new channels, we show that it's a np-hard too. In graph of a path, we prove there is a solution based on length of path,... 

Strategic Network Formation defines how and why networks take particular forms. In many networks, the relation between nodes is determined by the choice of the participating players involved, not by an arbitrary rule. A "strategic2 modeling of network requires defining a network’s costs and benefits and predicts how individual preferences become outcomes. In strategic network formation it is important to look at the overall social benefit and to see if networks that players create manage to be efficient for the society in general. The code review process also referred to as peer review, stands out as a tried and tested method in a large palette of applications to allow... 

Network traffic classification plays an important role in network monitoring and management. With the continuous development of network technology, traditional traffic classification methods face increasing limitations in accuracy, especially when dealing with encrypted traffic. Fortunately, deep neural networks offer an effective method for traffic classification due to their ability to learn the intrinsic features of data. In this study, we use the public ISCX network traffic dataset for our evaluations. We examine two general approaches. The first approach involves converting traffic into MNIST images and classifying classes using convolutional neural networks (CNNs), generating test... 

Software defect prediction is a significant and growing field in software development that focuses on identifying software modules prone to defects before the testing phase. The challenge of lacking historical data has shifted software defect prediction towards a cross-project approach. In this method, identifying defective software modules in a project (target project) is done using historical data collected from other projects (source project). The difference in data distribution between source and target projects is a major challenge in cross-project defect prediction. One way to overcome this challenge is the appropriate selection of the source project, which most existing studies... 

Path planning is essential for unmanned aerial vehicles (UAVs) to navigate complex and hazardous environments. Many algorithms have been proposed to solve this problem, but heuristic, meta-heuristic, and hybrid algorithms have recently gained attention due to their ability to find approximate solutions quickly. In this thesis, we propose using the Late Acceptance Hill Climbing (LAHC) algorithm for UAV path planning. We compare LAHC's performance to classical and state-of-the-art optimization algorithms, including simulated annealing (SA), grey wolf optimization (GWO), symbiotic organisms search (SOS), and simplified grey wolf optimizer-modified symbiotic organisms search (SGWO-MSOS).... 

Achieving sustainable production of high-quality food is a pressing issue for both developed and developing countries. With the rise in urbanization, climate change, and decreasing access to potable water, traditional agricultural methods are not suitable for meeting the future needs of the world. In this context, automated vertical farms are gaining attention. These farms allow us to minimize water consumption and pesticide use while signifcantly increasing the yield per square meter. To achieve sustainability in their operation, these farms must maximize effciency. A key challenge in this regard is minimizing the energy consumption of the elevators used in vertical farms. In this research,... 

Online question-answering communities, as reliable sources for exchanging experts' opinions, have specific rules to maintain their content quality. Due to their large number of users and posts, manual control and approval by administrators is not plausible, and these systems require solutions that are more scalable. The current dominant solution, i.e., the use of crowdsourcing and relying on user reports, has serious problems, including the slow speed of handling violations, the waste of time of users, and the discouraging feedback from the community towards new users. Although the automation of moderation actions via artificial intelligence methods would solve the existing problems, the...