Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 54608 (09)
- University: Sharif University of Technology
- Department: Civil Engineering
- Advisor(s): Kashani, Hamed
- Abstract:
- Optimal use of available resources for construction projects depends on proper decision-making on how to spend resources. Optimization of resource consumption should be done taking into account project revenue. In this regard, investors need a suitable mechanism to predict the selling price of the construction project and the resulting profit. Therefore, investors and stakeholders need a suitable model with the help of which they can make decisions such as choosing materials to guarantee the desired profit from the sale. This research will provide a framework for constructing models for estimating the price of residential units by considering some features of the building and some economic indicators such as the consumer price index. Based on this framework, first, the list of independent variables of the candidate for modeling was determined with the help of library studies. According to the determined variables, relevant data were collected for various residential units in the next step. The data were refined and pre-processed, and then the models were taught using various combinations of variables and machine learning algorithms. The performance of the models was evaluated using various indicators such as coefficient of determination, and appropriate models were selected based on performance. This model generally receives building specifications, including area, number of floors, type of materials, building location, etc., and predicts the estimated selling price according to market conditions. To show the mechanism of using the proposed framework, its application for the development of models for estimating the price of residential units in Tehran is described step by step. The basis of modeling in this research is machine learning methods, and the data required for modeling were collected in Tehran. The data of residential units in 22 districts of Tehran from the end of 1398 to the middle of 1400 were considered, which collected more than forty thousand data. This research aims to provide a framework that can be used to develop price estimation models at various points with the help of powerful machine learning algorithms. Users (investors, builders, buyers, and others) can use the resulting models to estimate prices and make decisions
- Keywords:
- Consumer Price Index ; Building ; Machine Learning ; Linear Regression ; Housing Unit ; Economic Index ; Price Estimation ; Housing Pricing
Digital Object List
-
محتواي کتاب
- view
Bookmark