Sharif Digital Repository

One of the influential factors affecting the future price trends of a stock is the public sentiment surrounding that particular stock. In recent years, researchers have employed Natural Language Processing (NLP) techniques to analyze textual data present on social networks, aiming to investigate public opinions. However, there has been limited attention given to validating the users expressing opinions concerning the stock market. Much of the opinions shared on social networks lack a thorough examination and analysis of the market, often being solely based on the author's sentiments. This research endeavors to validate active users on the social network 'X' (Twitter) by developing a... 

Bankruptcy is an important issue in the economy that can have extensive financial and social consequences on individuals and society. Timely warning to managers and providing analysis may prevent bankruptcy. Many studies have been conducted on the application and implementation of machine learning techniques to predict bankruptcy. Many bankruptcy prediction models produce incomprehensible outputs for the user. Therefore, they are called black box algorithms. Implementation of advanced models inevitably requires interpretability for users to understand the result and trust. Since most machine learning methods are "black box", explainable AI, which aims to provide explanations to users, has... 

Emergence of crypto-currency and blockchain technology revolutionize supply chain processes. In addition, customer needs for more information on products or services from origin to destination, highlights the necessity of transparency, originality and traceability in supply chains. This research is aimed to develop a blockchain based system ascertaining supply chain transparency and originality. To this aim, a joint pricing and closed-loop supply chain network design problem is selected as a good platform to implement it. Due to increasing concerns on environmental issues and maximizing job opportunities, sustainability is also considered in the proposed problem. To ascertain transparency... 

In this research, it has been tried to use deep learning methods and embedding vector, in addition to user-item data, from user side information such as age, gender, city, etc., and also for item information such as product name, product category, etc. can be used to better understand customer behavior patterns and provide a relatively rich recommender system. The proposed model in this research has two phases, the first phase tries to identify the user and item feature vector and form the user similarity matrix and the user-item correlation matrix. The outputs of phase one are used as inputs of phase two. In the second phase of the model, using these inputs, Top-N recommendation are... 

Research in the field of financial market prediction has always been an intriguing subject for academic researchers and stock traders, despite its associated complexities and challenges. Accurately forecasting stock prices and market indices is considered a complex task due to their nonlinear and dynamic nature, requiring analysis of intricate time series data. Over time, various models such as regression models, classification methods, statistical techniques, and artificial intelligence algorithms have been used to predict these variables. With the advancement of technology and the development of AI-based models, particularly machine learning models, along with the availability of vast... 

In the era of artificial intelligence, achieving high accuracy in machine learning models is crucial for their practical applications. This thesis presents a novel approach to improve the accuracy of machine learning models by learning to defer to a team of human experts. The primary goal of this work is to build upon and extend previous research, proposing a model that outperforms existing models in the literature. Inspired by the "Mixture of Experts" framework, we introduce a neural network-based allocation system responsible for assigning cases to each member of the team, which consists of a machine learning model and multiple human experts. The allocation system intelligently determines... 

Intensity-modulated radiation therapy is a well-known technique to treat cancer patients worldwide. A treatment plan in this technique requires decision-making for three main problems: selection of beam angles, intensity map calculation, and leaf sequencing. Previous works have investigated these problems sequentially. In this research, we present a new integrated framework for simultaneous decision-making of directions, intensities, and apertures shape, called direct angle and aperture optimization, and develop a mixed-integer nonlinear mathematical model for the problem. At first, the problem's single-objective model is established using the quadratic dose penalty function. After that,... 

The Vehicle Routing Problem (VRP) is one of the most widely studied topics in the field of Operations Research and has received tremendous attention, especially in recent years. This attention is driven by today’s customer expectations with respect to fast and reliable service. In (classic) static VRPs, all information is known at the time of decision making. However, the world is more dynamic now – and (naturally) so are the optimization problems. the advance of information and communication technologies, Global interconnectedness, urbanization, and increased service orientation raise the need for anticipatory real-time decision making. A striking example is Logistic Service Providers... 

What people write about their experience on web pages or social media about a product they have used or a service they have received can influence the reputation and the popularity of a certain brand with a great deal. If the reviews that exist about a product or a service of a certain company are mainly positive, it can increase the profit and improve the image of the company. On the other hand, mostly negative reviews can decrease the profit and destroy a company's image irreversibly. Unfortunately, because of this great influence that online reviews have over general public's decision to use a a product or a service of a brand, some companies hire people to write undeserving positive... 

This study proposes a deep reinforcement learning approach to solve a perishable inventory allocation problem in a two-echelon supply chain. The inventory allocation problem is studied considering the stochastic nature of demand and supply. The examined supply chain includes two retailers and one distribution center (DC) under a vendor-managed inventory (VMI) system. This research aims to minimize the wastages and shortages occurring at the retailer's sites in the examined supply chain. With regard to continuous action space in the considered inventory allocation problem, the Advantage Actor-Critic algorithm is implemented to solve the problem. Numerical experiments are implemented on... 

In the last decade, emergency department (ED) overcrowding has become a national crisis for the US healthcare system. Increasing mortality rates, decreasing quality of care, financial losses due to walkouts, and ambulance diversion are some of the consequences of ED overcrowding. Given the increasing demand in terms of ambulance utilization which we can see an instance of it in the COVID-19 pandemic, being able to allocate service requests to EDs efficiently, becomes a key function of emergency medical services. in this investigation, an algorithm of deep reinforcement learning called deep Q-learning is used to address this problem and to assign ambulances to ED's appropriately. under... 

Today, machine learning has many applications in various industries, and healthcare is not an exception. Machine learning algorithms are used for medical diagnosis, make predictions about patients’ future health, newly-discovered treatment effect on patients prediction, drug recommendation system, build risk models and survival estimators and health risk prediction models. One of the topics that has received less attention in the world, especially in Iran, is the prediction of the surgery duration. This is very important because operating rooms in hospitals are the primary source of hospital revenue; We also need to predict the duration of surgery as accurately as possible in order to... 

Ⅾue to the absenⅽe of the ⅼabeⅼs or so−ⅽaⅼⅼeⅾ target variabⅼe، ⅽⅼustering vaⅼiⅾation، ⅾespite ⅽⅼassifiⅽation, is not that straightforward. So، the ⅽⅼuster evaⅼuation is a ⅽhaⅼⅼenging task both in researⅽh projeⅽts anⅾ appⅼiⅽations. Whiⅼe ⅿany ⅽⅼustering vaⅼiⅾity inⅾiⅽes are aⅾⅾresseⅾ in the ⅼiterature, ⅿost of theⅿ, even those wiⅾeⅼy useⅾ in the appⅼiⅽation, ⅽannot hanⅾⅼe arbitrary shapes. In this paper, a noveⅼ ⅽⅼustering vaⅼiⅾity inⅾex is proposeⅾ, whiⅽh is ⅿuⅽh ⅿore powerfuⅼ in ⅽapturing the ⅾata’s reaⅼ struⅽture anⅾ ⅾeaⅼing with arbitrary shapes. An aⅼⅿost noveⅼ separation ⅿeasure is proposeⅾ to represent the signifiⅽant or insignifiⅽant separation regarⅾing the ⅽⅼuster’s struⅽture... 

Identifying anomalous events is one of the vital topics in research as it often leads to the detection of actionable and critical information such as intrusions, faults, and system failures. With its importance, there has been a substantial body of work for network anomaly detection using supervised and unsupervised machine learning techniques with their own strengths and weaknesses. In this work, we take advantage of both worlds of unsupervised and supervised learning methods. The basic process model we present in this paper includes (i) clustering the training data set to create referential labels, (ii) building a supervised learning model with the automatically produced labels, and (iii)... 

With the development of online two-sided platforms and increasing competition between these companies, issues such as customer targeting or recommendation systems have become more important to organizations. So far, various tools have been used for this purpose, but one of the most effective methods is the data analytics based on the stored data, through which personalized promotions can be automatically sent to the customers by implementing optimization models and algorithms. In this research, we present a model that re-adjust the commissions received from drivers based on detecting hidden patterns in their behavior in order to maximize the company's profit and then offer a suitable... 

Unstructured data is an important resource in data mining which In spite of their large volume, they haven’t been analyzed so much. Natural language data are a typical kind of unstructured data which humans can easily understand them but normally it is not possible for machines to process these kind of data. To make these data usable for prediction, pre-processing is required to prepare them for feeding into machine learning algorithms. Therefore, feature extraction is needed for texts in order to make presentative features from them that can unveil the hidden pattern. In this study, in addition to the variables that extracted from the technical indicators, the texts from telegram channels... 

Today, with the development of financial and economic sciences and the increasing volume of financial data, it is necessary to process and analyze this field more accurately with up-to-date tools. On the other hand, by the significant growth of the use of machines and computers for analysis and forecasting purposes, their importance and application have been well defined. Therefore, this research is considered to provide a more efficient method by processing historical data and analyzing them using data mining techniques. The results of this study can be provided to experts in this field as an effective method. Therefore, in this research, a new method based on the selection of required... 

Over the past few decades, with the exponential increase in data volume, scientists and researchers have tried to discover relationships and algorithms for productivity and find useful information from this amount of data in various fields. Their efforts in data analysis have led to the development of algorithms in the big data field. The result of researchers' working in multiple fields has come to aid the people of science and technology. Among the most important of these areas, we can mention the health and medical sectors, financial sectors, services, manufacturing sectors, etc. The purpose of this study is to enter the financial industry and use data mining tools. One of the newest and... 

The research in this study aims to analyze data in healthcare, especially the diagnosis of several diseases caused by heart failure. Analyzing and analyzing this data can lead to the discovery of relationships and patterns that can play an important role in the decision-making process of relevant officials in any field. Today, medical data around the world is stored in large volumes for future research. Various infrastructures and software have been set up in many health centers and research centers affiliated with those organizations.In this research, the general process of work is such that the data related to the heart sounds, which are in the four broad categories of S1 to S4, are... 

Investors' sentiment toward a coin is very vital in cryptocurrencies' markets. If traders do not invest in a coin, its price will decline; therefore, it is essential for investors to take others' sentiment into account when they want to buy or sell a coin. Text mining on social media is one technique that can help traders understand others' opinions about the coin they are trading. Hence, in this research, we try to predict the top three cryptocurrencies, Bitcoin, Etheruem, and Litecoin, price movement for the next day based on Twitter posts and News title using text mining. In this research, we found out that considering all tweets can reduce our model's accuracy, and for better accuracy,...