Sharif Digital Repository

When the keywords in academic articles are collected and accumulated to achieve a certain amount, they will display a network property. This property can help us to get the hotspot graph of science by using the keywords network, which is directly related to the subject of the paper. Also by analyzing this network, it is possible to forecast the future science hotspots.By analyzing keywords network from complex networks point of view, we are able to predict the future hotspots based on current trends, previous growth patterns that happened or with the power of different centrality measures like eigenvector etc. As an example, when a new keyword appears in a paper for the first time, a complex... 

Poor decision making in health care has always had irreparable consequences for society. Also, expensive medical tests cause lots of problems for patients. A huge amount of data is produced daily by hospitals, which unfortunately are not used to improve decision making and predicting disease. Data mining can be an appropriate tool for extracting knowledge from a huge amount of data by using a variety of techniques such as prediction. The leading cause of death in the world is heart disease so this study has been designed to predict the incidence of that. Regarding the literature review, the Naïve Bayes method had predicted heart disease accurately. According to the specialist's opinion, some... 

Today with the development of technology in medical science, there is a need to develop new methods to analyze and process the medical images. Furthermore, increasing use of machines and computers to accomplish prediction goals delineates that these tools had promising results. Because of all the above, this research focuses on processing and analyzing medical images with using data mining tools in order to identify MS and tumor disease which have been ubiquitous in last decades, fast and meticulous. To do so, we introduce a new clustering algorithm based on the modularity measure of graph networks as well as a new machine learning algorithm based on Kalman filter for Tensor-based data.... 

This research uses the Comovement between electronic payment data with GDP growth rate and private consumption growth rate to nowcast these two important macroeconomics variables. We use point of sale data set from the beginning of 1392 to the end of autumn of 1395 in monthly frequency in mixed data sampling model to use the Comovement for Nowcasting. Besides, we use two other data set, sale of industrial electricity and sale of agricultural electricity. All of this data set gives a real time sense of macroeconomics. They are also free of mistake. We show that using of this data set in the frame of MIDAS can decrease the root mean square forecast error (RMSFE) relative to auto regressive... 

Estimating the remaining useful life (RUL) of critical assets is an essential task in system health management. It also has a crucial role in the optimization of maintenance scheduling and decision making about the future of the asset. On the other hand, since the rolling element bearings are widely used in the rotating machinery, estimation of their RUL is considered as remarkable progress in improving the reliability of the whole system. In this way, various models have been introduced to predict the RUL of rolling element bearings, which their proper functionality is affected by the underlying assumptions in the model structure. Because of the presence of several uncertainties and... 

One of the most important problems in online learning is a prediction with expert advice. In each step we make our prediction not only based on previous observation but also use expert information. In this thesis, we study the different well-known algorithms of expert advice and generalize problems when data arrival is in the two-dimensional grid. regret is a well-studied concept to evaluate online learning algorithm. online algorithm when data arrive consecutively in T time step has regret O (√(T)) . regret in two-dimensional grid with T row and P column is O(T√(P)).
2010 MSC: 68Q32 ; 68T05 ; 90C27  

The importance of transfer learning in image-related problems comes from its many advantages that are sometimes undeniable. Previous researches have well shown the success of transfer learning in this area using deep neural networks. However, transfer learning for time series data has not yet been done in a conventional and automated manner. The main reason for avoiding transfer learning in this domain relates to the dynamic and stochastic nature of the time series, where they show a time-varying behavior. Previous experiments have shown that transfer learning between two heterogeneous time series could harm the forecasting accuracy of a model. Therefore, in this thesis, we aim to explore... 

One function of futures markets is to signal about future spot prices. This role is accommodated by creating a way for future tradings. In futures markets, different types of investors, including speculators, arbitrators, and hedgers, trade based on their expectations of the future. The better the agents are in pricing, the more significant this role of financial markets. Therefore, it is possible that futures prices contain information that helps to predict maturity spot prices. This study examines the performance of using gold coin futures prices in forecasting with the help of data from December 2008 to July 2016. Studying the cointegration relationships between futures and maturity... 

Making money and maintaining the value of assets has always been one of the most important concerns of people. Real estate is one of the essential human needs, but it is also considered an investment tool for individuals. In addition to individuals in a family, various groups and organizations such as policymakers, analysts, banks and financial institutions, taxpayers, and real estate investors are directly or indirectly affected by the dynamic characteristic of the housing market. Therefore, forecasting the exact amount of housing value in the future is very important. Factors that can improve this forecasting's accuracy include considering the relationship between housing value and... 

Recently, data scientists have been paying much attention to sports. Many researches have been done in this field, using data mining and machine learning techniques. The following research aims to predict the results of football matches, which consists of two general approaches. For the first and second approaches, we used video game data and match statistics, respectively. In both approaches, it was tried to predict not only the final result (win, draw, or loss) but also the final goal difference. In the first approach, the home team victory was predicted by 73% accuracy, the draw by 75.4%, and the home team defeat by 73.7%. Nevertheless, in the second approach, the home team victory was... 

In a real-time routing problem, a sequence of requests emerge overtime in a metric and should be served by $k$ moving agents (i.e. the servers) in order to minimize a certain cost function. If the cost function is the average of completion time of the requests, the problem is named \textit{online $k$-traveling repairman problem}. An online algorithm with lookahead would be aware about next arising requests in a specific time-window. We present deterministic and randomized algorithms with competitive ratios $5.829-\delta$ and $3.873-\delta$ against and oblivious adversary, where $\delta$ depends on server's lookahead, specification of metric and last request's release time. We the show... 

Population growth and the escalating use of electric devices have led to a surge in electricity demand. However, the construction of adequate infrastructure to meet this additional demand is highly expensive. To address this issue that sometimes causes power outages, dynamic pricing for electricity, a demand management technique, has gained significant importance. By employing this method, prices can be adjusted on an hourly basis, enabling consumers to regulate their consumption in response to these price fluctuations. The prices can be determined based on different objective functions. This research aims to develop a comprehensive framework for determining the optimal hourly electricity... 

Generally, nowadays, machine learning methods are used in many different areas for their superiority over other methods of prediction. Although being a tough task, stock market prediction with machine learning approaches is being spread due to satisfying results published ever yday. Machine learning methods usually use varied kinds of data ,including structured data such as market data, technical indicators and some fundamental data as well as unstructured one entailing text and graph data in order to enhance their predictability capacity. In this thesis we aimed to find out more about the importance and the contribution of structured data in prediction and we tried to attain a framework... 

Groundwater beneath the cities is becoming an important and valuable resource. Conjunctive use of surface and groundwater is likely to become increasingly more common as urban population grows by time. Therefore, one important requirement for urban water management planning is forecasting the groundwater level fluctuations. Unfortunately less experience and information is available to evaluate the fluctuations of groundwater level in urban environment compare to the natural systems, also different processes (sources) are involved in an urban water cycle, which all together make it more complicated to study. Similar to many other megacities, there is a serious lack of hydrogeological and... 

today's financial markets such as stock market are more attractive and important position and wealth are considered income and therefore attracts many people have. But the other hand, activity in these markets requires a high risk of admission. The point that is important is that the risk of investing in these markets can be predicted to some extent with the trend of stocks and securities can be controlled. Time series trend of stock prices and non-static characters is excited. But analysis of such behavior is impossible, i.e., reliance on sophisticated tools and of course accept the possibility of an error can be predicted price to pay. Synthetic models of artificial intelligence today, due... 

The Urmia Lake in the north west of Iran and one of the most important habitats in the world is in danger of drying. Drought of recent years, increasing of temperature and evaporation and also construction of several dams in Urmia Lake basin can be considered as the main factors of decreasing the Lake level. Simultaneous forecasts of lake level and inflow streams help us to make better decisions for allocating and releasing enough water for environmental demands such as Urmia Lake. This study aims to determine relationships between historical information of basin with streamflow of Ajichai and Urmia Lake level, and use them for predicting the further conditions. In this process, streamflow... 

This research proposes a novel hybrid method for stock price index forecasting relying on technical analysis, the fundamental analysis of capital market, neural networks technology, and fuzzy logic. The new method shows a good performance comparing complicated and time-consuming forecasting methods. Several factors influence stock price; for example, an important one is the previous trend of stock price. Fortunately, by developing technical analysis and introducing the various indices of this method the maximum use of historical data is made in order to forecast future prices. Another kind of factors is macro-economic variables that their influence on the long-term trends of the stock... 

The water demand forecasting is an important activity for successful planning, utilization and operation of urban water supply and distribution systems. The population growth result in water consumption growth, also the restriction of water resources lead to pay more attention to the water demand management. The unexpected droughts, financial crises, over-use of water resources, or inessential infrastructure development are the outcome of poor water demand prediction and inflexible water resource management. This research is addressed the daily short-term (two week ahead), weekly medium-term (six months ahead), and monthly long-term (two years ahead) water demand. The dynamic artificial... 

In this Thesis, a multi period portfolio optimization consisting stocks, gold and risk free asset is considered, in which periodical reinvestment and withdrawing is possible. Maximizing the net present value of investor’s cash flow is the objective. Due to the existence of uncertain parameters, two robust counterpart models are developed. In the first model, a conservative robust model is presented to generate feasible solution in all cases. In the second one, the conservative degree of investor is adjustable to control the risk of the model by investor appropriately. For evaluating the proposed models, the data of 5 well known stocks of Tehran market and gold prices are gathered. By using... 

In this thesis, we examine different forecasting methods to predict volatility in financial markets. Tehran Exchange Price Index (TEPIX) is adapted to forecast in short and long term periods. TEPIX is the most important index in Tehran Stock Market which is officially reported daily. Autoregressive Integrated Moveing Average models (ARIMA), Generalaized Autoregressive Heteroskedastic models (GARCH) and Artificial Neural Networks (ANN) are used for forecasting TEPIX. Spectral Analysis is also regarded as a completely new approach in financial mathematics to forecast TEPIX in short and long term periods. We consider different criteria to compare the performance of different methods of...