Sharif Digital Repository

In today’s competitive world, using an efficient forecast method is necessity for companies. To now, many forecast methods have been developed, but many of them have not an expected efficiency. In this research, we develop a new hybrid forecast method with application of forecasting retails demand. The hybrid method is the combination of ARIMA method and neural networks. To test the efficiency of the method we use the 96 weeks data of plastic containers demand. We also comprise the hybrid method with other forecast methods including naïve method, ARIMA method and neural network method by applying root mean square error and mean absolute percentage error indexes. In the case of plastic... 

Failure of mechanical parts in the industry lead to a larger system downtime and even imposing economic losses to the factory. For this Purpose, for many years, researchers have been trying to find ways to predict early failure and to prevent losses from occurring. Creation of new sciences like artificial intelligence, helped researchers in this field.In the current study, using experimental data of a set of bearings that have been tested and recorded in the Intelligent Systems Research Center, A new approach with sufficient accuracy is presented for the prediction algorithm. Among the features extracted, three features of entropy, root mean square and maximum are the most appropriate... 

P/E ratio is studied in four levels in this study:
1)Macroeconomics level
2)Capital market level
3)Industry level
4)Company level
The first level studies effects of macroeconomics variables on P/E ratio. At this level we use variables such as economic growth, inflation, exchange rate, and etc.The next level uses capital market variables such as market volume, and IPO information.The third level that we study in this research is industry level. Stocks of an industry generally behave similar, because they have common advantages and disadvantages, thus industry is an effective factor on P/E ratio.The last level studies financial statements and internal features of a... 

This thesis firstly studies the parameters affecting P/E ratio. These parameters vary from Macroeconomics level, Economic growth and Inflation, to company level. Then this study deploys Neural Networks to predict magnitude of P/E and change direction of P/E ratio. To increase accuracy, thesis uses three different method of normalizing for Input data. Finally, results are compared to results of regression method  

Sign language is recognized by considering the combination of the hand gesture, orientation, location and patterns of hand and arm movement. These complex interactions make the character and word recognition very challenging. In this paper, with the aid of sensory gloves and multiple inertial measurement units (IMUs) we measure the shoulder and elbow joint trajectories and hand gesture to train the discriminant functions to recognize the words intended and represented by sign language. For recognition of hand gesture, the Naive Baysian classifier was used while for Eulerian angles a new time series similarity measures were computed. Different aggregation method was used to integrate temporal... 

Shortage in fossil fuels resources together with the pollution concerns have caused a systematic change in power system planners and decision makers policies to renewable energies as an alternative to produce electrical energy. However, some intrinsic features of the wind energy overshadow its profitability. Inability to predict the wind speed changes and consequently the output level of wind turbines, being an uncontrollable generation unit, and also being an intermittent unit can be accounted as the main attributes of renewable-based units. Taking into account these features, one can conclude that new challenges can be brought into existence in planning and operation issues of... 

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... 

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... 

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... 

Action recognition has a lot of applications in everyday human life. In the past, the researchers concentrated on using RGB frames, but since the advent of 3-dimensional sensors such as Kinect, 3D action recognition drew researchers' attention. Kinect can extract the joints of the body in action as time series. One of the main challenges of action recognition is that different individuals perform an action with various styles and speeds. Hence, the conventional methods such as calculating Euclidean distance seem inappropriate for this task. One solution is to use the techniques such as DTW, which aims to temporal aligning of the sequences. The DTW is not a metric distance; hence, in this... 

Artificial neural networks are mathematical models inspired by the nervous system and brain. The types and applications of these networks are very widespread nowadays, and it seems that they can be used to track the signals well and estimate the data of the next. In this research, we try to present a model that can predict the future of the trend of noisy signals that have unpredictable behavior, or in other words, chaotic signals. Such research is also widely used in the medical sciences, including the diagnosis of epileptic seizures or heart attacks. In this research, a study with high volatility financial data has been done as an example on this issue and the proposed model tries to be... 

Predicting oil prices, especially in exporting countries, will help governments in the policy-making process by obtaining a reliable estimate of oil revenues. The existence of a complex mechanism governing the process of oil price formation has reduced the efficiency of linear models in forecasting and led researchers to use nonlinear intelligent systems to predict oil prices. In this study, after a detailed study of the structure of artificial neural network, two models of neural network GMDH and MLP and ARIMA method have been used to predict oil price. There are important factors in the prediction process with neural networks, and if all these factors are selected correctly; One can expect... 

Learning is an essential mechanism for the survival of living things. There are different types of learning, and value learning is among the most important types. A child learns that water resolves the thirst need by repeatedly experiencing this situation. Eventually, the value of water, which has been valueless before that, increases gradually in his mind. How this concept is encoded in the brain? previous works reveal the role of different neurons and regions that are relevant to value learning. However, population analysis and dynamic modeling are less considered. Moreover, the links between different brain regions are unknown.Finding the relationship between two relevant regions of the... 

Real-time travel time estimation is a major requirement in many transportation related systems. One of the main challeges is to estimate the traffic speed and then forecast it for a short time. A valuable data source for this task is instant location of moving cars that is captured using global positioning system (GPS) and sent through internet in online manner. The main problem is that the resulting traffic data is severely sparse and also contains a lot of noise. Previous researchs on this type of data are mostly based on matrix or tensor factorization. In this work it is shown that despite the large fraction of missing value it is possible to use neural network for this problem with some... 

Water shortages resulting from macro-environmental climate changes as well as local inefficient agricultural practices and dam constructions activities have resulted in the gradual reduction of water level in Urmia Lake, located in the northwest of Iran. As such, restoration efforts were initiated to prevent further adverse impacts exacerbating the conditions and creating secondary problems such as regional salt dust generation and dispersion, resulting in health issues for the greater area population in the neighboring vicinities. The utilization of advanced forecast modeling based on deep learning algorithms can assist the authorities to manage better multi-dimensional issues affecting the... 

This study proposes a method to predict the remaining useful life (RUL) of the rolling element bearings (REBs) by forecasting the future trend of the peak of the acceleration signal. It is also employed to determine an appropriate time interval between the measurements of REBs vibration to reduce the error of forecasting and avoid collecting too much data in addition to increasing the reliability. In the first step, in order to achieve better results, the history of the acceleration peak is transformed into a stationary space before using the long short-term memory (LSTM) model to make it normally distributed and stationary. Then, LSTM forecasts the future trend of the stationary time series... 

Performing analysis and forecasting is crucial. Better forecasting will lead to better decisions. One method for predicting the future is time series analysis. In reality, it is common for an intervention to occur and alter the characteristics of a time series. In recent years, interrupted time series analysis has been receiving a lot of attention. A new forecasting method for interrupted time series has been developed in this study. This is a system dynamics-based approach. At every stage of the approach, system thinking is incorporated. In order to model the effects of a given intervention, common modes of behavior in dynamic systems are used. Furthermore, control theory has been used to... 

Nowadays, various policies are adopted by transportation managers and planners. These policies aim to improve system performance, reduce user costs, control and reduce air pollution, reduce noise pollution, and ultimately reduce congestion. A set of these policies in the form of transportation demand management is presented in the literature. A common way to find the effect of a policy on user behavior is to use questionnaires. Other causal inference models have been proposed in disciplines such as statistics, political science, marketing science, epidemiology, and psychology. The purpose of these models is to find the causal effect of an intervention (treatment) on a system. These studies... 

This work presents a modification of deep neural networks for time series analysis. We used kernel layer(s), as a novel approach, at the beginning of the common deep neural networks. These kernels learn based on dynamic time warping (DTW). In each kernel, DTW is calculated between the kernel value and a part of input time series or a part of last layer output (if the kernel is not in the first layer). DTW also gives an alignment path for the input series. This alignment path is used to defining a loss function with the goal of getting better alignment (lower DTW distance) between the kernel and the other input. Besides getting better accuracy on the examined datasets, the other achievement... 

Using revenue management in any industry can increase the profit. In aviation industries, due to the huge number of requests and travels for each airline, a revenue management system can lead to a good profit for the airlines. The first step in revenue management system is predicting the demand.In this article two models are developed using time series techniques, based on the information taken from one of the Iranian airlines in Tehran-Mashhad fly route.The first model is developed using ARIMA and seasonal-ARIMA models and the second one is based on the demand and price history, price in the day of prediction and the ARIMA model. The second model which is a combination of price, prior price...