Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/68181
Title: Housing price prediction using neural networks
Authors: Liu, Zhaoying
Keywords: DRNTU::Engineering::Electrical and electronic engineering
Issue Date: 2016
Abstract: Artificial Neural Network (ANN) is inspired and developed by modern neuroscience, which aims at reflecting the structural and functional characteristics of the human brain by building abstract mathematical models. It is a simple simulation of the complex biological neural network to develop an artificial system that could perform "intelligent" tasks similar to the human brain. The neural network system is composed of interconnected neurons which exchange messages through connections with numeric weights that can be adjusted by learning. There are more than 40 different kinds of neural network models, the most common ones are Multilayer Perceptron (MLP), Hopfield net, Boltzmann Machine, Backpropagation (BP) net, etc. Among all the various capabilities of neural-net computing, supervised learning is the most prominent one that can be implemented using multi-layer feed forward net. The BP neural network regarded as the typical representative of feed forward net has its distinct disadvantages like slow learning and convergence rate and nonideal adaptive ability. As a result, the high-order functional-link net theory was brought forward along with the BP algorithm, which does not only enhance the learning speed but also has an unexpected function in reducing the learning algorithm and network topology construction. And one particular approach is the random vector implementation of the functional-link net. Nowadays neural networks have achieved great success in the application of pattern recognition, image processing, combination optimization, decision making, classification, and prediction. With the soaring housing price and pressure of purchasing houses, housing price has become a universal attention for the public. Housing price forecasting is a very good reference for the real estate investment and government's macro-control. In this project, the widely used BP network as well as random vector functional-link (RVFL) network are implemented in housing price prediction.
URI: http://hdl.handle.net/10356/68181
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
FYP_Report_Final.pdf
  Restricted Access
3.44 MBAdobe PDFView/Open

Page view(s)

155
Updated on May 14, 2021

Download(s) 50

43
Updated on May 14, 2021

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.