جستجوی مقالات مرتبط با کلیدواژه « انتخاب ویژگی » در نشریات گروه « اقتصاد »

Introduction :

Smart cities form the future of cities and the age of information revolution will sooner or later lead cities to become smart. The smart city has become very important in today’s technological world and is followed as an important model in advanced countries, but in order to realize the smart city, there are requirements and essentials, as well as components, which their realization is inevitable. Therefore, cities should have the technical preparation to become smart or provide the necessary platform. Therefore, two questions are raised what are the technical factors that are effective for the implementation of urban smart projects and what is the relationship between these factors?

The current research is applied in terms of purpose and descriptive in terms of data collection. The type of research is a mixed qualitative and quantitative type. Data collection is done at both field and library levels. In order to investigate the research gap and discover innovation, the library method is used, while using this method, theoretical foundations, and research background are extracted, but to identify the technical factors affecting the implementation of the smart city, interviews with experts are used. In addition, to extract data, governmental and archival databases are used. The data collection tool in the current research includes an interview and a Delphi questionnaire with a scale of 20, and each of the factors extracted in the previous step is included as an item in this question. The validity of this questionnaire is checked using the opinion of professors and its reliability is checked using Cronbach’s alpha test. The statistical population of this research includes all experts and activists in the field of smart city, who must have been working in the relevant field for at least 10 years, and the sampling method is snowball. Excel software was used in the Delphi and classification stage, and MATLAB software was used in the feature selection section.

By using the content analysis technique, factors affecting the smart city are extracted. These factors were extracted by using interviews with experts. After determining the factors, the extracted technical factors were reviewed and refined in the Delphi stage, and the factors that were not agreed upon in the first three stages were eliminated. Out of 22 technical factors, 4 factors were removed due to the lack of consensus and the existence of discrepancies in the three Delphi steps. Cronbach’s alpha test statistic for all variables is higher than 0.6, which shows that all variables are at the desired level of reliability, and therefore the reliability of the questionnaire is confirmed. Then, using the genetic meta-heuristic algorithm technique, which is one of the feature selection methods, the final screening of the input variables was done. The classification stage is considered the final stage of the current analysis. In the previous stages, variables were identified, screened, and refined. However, at this stage, using the interpretive structural modeling method, the leveling of the criteria derived from Delphi methods and the selection of characteristics based on effectiveness and affectability are discussed, and according to the final model of the research, the sixth level, i.e. digital infrastructure criteria, is the most effective. It is a criterion that directly affects the fifth level criterion, which is the development of Internet infrastructure. The first level, which includes the criterion of intelligent driver assistance systems and building warning systems, is the most effective criteria.

In the near future, the smart city will emerge as an important paradigm in the field of urban management and urban planning, while in the developed world this concept is more institutionalized than in the developing world, so research in this field and identifying the factors affecting it It is considered an inevitable necessity. The final model of technical factors includes six levels. The sixth level, i.e. the digital infrastructure criterion, is the most influential criterion, which directly affects the fifth level criterion, i.e. the development of the Internet infrastructure. Also, technical factors were investigated in terms of influence and dependence, based on which, digital infrastructure criteria and Internet infrastructure development are independent. These variables have low dependence and high directivity, in other words, high influence and low influence are the characteristics of these variables. The criteria of intelligent driver assistance systems and building warning systems are also of the dependent type, which have strong dependence and weak guidance. These variables have high influence and little influence on the system. The rest of the criteria are of the interface type, these variables have high dependence and high guiding power, in other words, the effectiveness and affectability of these criteria are very high, and any small change in these variables causes fundamental changes in the system. Therefore, the results of this research show that digital infrastructure is the most effective technical factor in the implementation of smart city plans, and investing in this sector is very important in the development of smart city plans, which will have a significant impact on the lives of citizens and the urban economy.

In capital markets, stock price forecasting is affected by variety of factors such as political and economic condition and behavior of investors. Determining all effective factors and level of their effectiveness on stock market is very challenging even with technical and knowledge-based analysis by experts. Hence, investors have encountered challenge, doubt and fault in order to invest with minimum risk. In order to reduce cost and raise the profit of investment, determining effective factors and suitable time for sailing and purchase is one of the important problems that every shareholder or investor in stock market should consider. To reach this goal, a variety of approaches have been introduced, which are often intelligent, statistical, and hybrid. These approaches are mostly used to predict the stock price time series. Our proposed algorithm is hybrid and involves two stages: preprocessing and predictor. The preprocessing stage involves three steps: missing value, normalization and feature selection. Since there are many features in used datasets, genetic algorithm (GA) is used as the feature selection algorithm. In order to intelligent capability of Fuzzy Neural Network (FNN), this network with two structures (Mamdani and Sugeno) is used as a stock price prediction in second stage. This network is capable of extracting fuzzy rules automatically. Back propagation algorithm (gradient decent) is used for adapting all the parameters.
Our algorithm is evaluated on ten datasets with seven features obtained from ten different companies. By comparing the simulation results of the simple and hybrid FNN network, we found that the lack of suitable feature selection algorithm will lead to high computational cost, and in many instances the hybrid algorithm outperforms the simple FNN. This results demonstrate the superiority of the hybrid FNN to the simple one. In general, since the number of Sugeno tuning parameters are more than Mamdani, its performance is better than mamdani. Moreover, our algorithm is comparable to the maximum precision rates of other approaches.