Analyzing Massive Data Sets: An Adaptive Fuzzy Neural Approach for Prediction, with a Real Estate Illustration

Guan, Jian; Shi, Donghui; Zurada, Jozef; Levitan, Alan S.

doi:10.1080/10919392.2014.866505

Cited by 32 publications

(28 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The value of any building or plot of land belongs to a rich network where decisions about and perceptions of neighboring properties influence the final market value. Guan, Shi, Zurada, & Levitan (2014) compared traditional MRA techniques to alternative data mining techniques resulting in mixed results. However, as Helbich, Jochem, Mcke, & Hfle (2013) state, hedonic pricing models can be improved in two primary ways: through novel estimation techniques, and by ancillary structural, locational, and neighborhood variables.…”

Section: Predicting Gentrification Using Machine Learningmentioning

confidence: 99%

“…Antipov & Pokryshevskaya (2012) came to a similar conclusion about the superiority of Random Forest for real estate valuation after comparing 10 algorithms: multiple regression, CHAID, exhaustive CHAID, CART, 2 types of k-nearest neighbors, multilayer perceptron artificial neural network, radial basis functional neural network, boosted trees and finally Random Forest. Guan et al (2014) compared three different approaches to defining spatial neighbors: a simple radius technique, a k-nearest neighbors technique using only distance and a k-nearest neighbors technique using all attributes. Interestingly, the location-only KNN models performed best, although by a slight margin.…”

Section: Predicting Gentrification Using Machine Learningmentioning

confidence: 99%

See 1 more Smart Citation

The spatially conscious machine learning model

Kiely

Bastian

2019

Statistical Analysis

View full text Add to dashboard Cite

Successfully predicting gentrification could have many social and commercial applications; however, real estate sales are difficult to predict because they belong to a chaotic system comprised of intrinsic and extrinsic characteristics, perceived value, and market speculation. Using New York City real estate as our subject, we combine modern techniques of data science and machine learning with traditional spatial analysis to create robust real estate prediction models for both classification and regression tasks. We compare several cutting edge machine learning algorithms across spatial, semi-spatial and non-spatial feature engineering techniques, and we empirically show that spatially-conscious machine learning models outperform non-spatial models when married with advanced prediction techniques such as feed-forward artificial neural networks and gradient boosting machine models.

show abstract

Section: Predicting Gentrification Using Machine Learningmentioning

confidence: 99%

Section: Predicting Gentrification Using Machine Learningmentioning

confidence: 99%

The spatially conscious machine learning model

Kiely

Bastian

2019

Statistical Analysis

View full text Add to dashboard Cite

show abstract

“…Other works rely on expert systems based on fuzzy logic because of the similarity between this technique and the human approach to decision making [7]. For example, Guan et al [8] have used adaptive neuro-fuzzy inference system (ANFIS) for real estate appraisal. To do so, they have used data from 20,192 sales records in a mid-western city of the US between 2003 and 2007, which were finally reduced to 16,523 after manual curation.…”

Section: State Of the Artmentioning

confidence: 99%

Identifying Real Estate Opportunities Using Machine Learning

et al. 2018

View full text Add to dashboard Cite

The real estate market is exposed to many fluctuations in prices because of existing correlations with many variables, some of which cannot be controlled or might even be unknown. Housing prices can increase rapidly (or in some cases, also drop very fast), yet the numerous listings available online where houses are sold or rented are not likely to be updated that often. In some cases, individuals interested in selling a house (or apartment) might include it in some online listing, and forget about updating the price. In other cases, some individuals might be interested in deliberately setting a price below the market price in order to sell the home faster, for various reasons. In this paper, we aim at developing a machine learning application that identifies opportunities in the real estate market in real time, i.e., houses that are listed with a price substantially below the market price. This program can be useful for investors interested in the housing market. We have focused in a use case considering real estate assets located in the Salamanca district in Madrid (Spain) and listed in the most relevant Spanish online site for home sales and rentals. The application is formally implemented as a regression problem that tries to estimate the market price of a house given features retrieved from public online listings. For building this application, we have performed a feature engineering stage in order to discover relevant features that allows for attaining a high predictive performance. Several machine learning algorithms have been tested, including regression trees, k-nearest neighbors, support vector machines and neural networks, identifying advantages and handicaps of each of them.

show abstract

“…При построении моделей рынка недвижимости обычно исполь-зуют две разновидности информационного подхода: регрессионную [2, 4-8] и нейросетевую [9][10][11][12][13][14][15][16][17][18][19]. В качестве входных (независимых) пе-ременных, как правило, учитывают строительно-эксплуатационные параметры, такие как тип объекта, место его расположения, площадь, количество комнат, этажей, наличие балкона, лоджии, парковки и др.…”

Section: регрессионный и нейросетевой подходы к построению моделейunclassified

“…Следует отметить, что как регрессионные, так и нейросетевые мо-дели [2,[4][5][6][7][8][9][10][11][12][13][14][15][16][17]19], учитывающие одни только строительно-эксплуата-ционные факторы, традиционно используются для массовой оценки недвижимости в ряде стран с развитой и достаточно стабильной эконо-микой. В развивающихся же странах, включая Россию, рынок недвижи-мости подвержен влиянию быстроменяющихся мега-, макро-и мезоэко-номических факторов.…”

Section: традиционные и комплексные моделиunclassified

Discussion of Data Mining, Mass Appraisal and Management of Real Estate Regional Market

Alekseev¹,

Kharitonov²,

Iasnitskii³

2017

AMCS

View full text Add to dashboard Cite

НЕДВИЖИМОСТИ РЕГИОНОВ РОССИИОбсуждаются преимущества использования информационного подхода к моделированию регионального рынка жилой недвижимости. На примере г. Екатеринбурга показана необходимость учета влияния мега-, макро-и мезоэкономических факторов при построении компьютерной модели рынка, пригодной как для анализа и прогнозирования рыночной стоимости жилых объектов, так и для массовой оценки кадастровой стоимости, лежащей в основе эффективной системы налогооб-ложения, направленной на справедливое распределение налоговой нагрузки. Помимо этого, ин-теллектуальный анализ моделей «что если» позволяет видеть, как изменения ключевых ценообра-зующих факторов влияют на рыночную стоимость объектов, что может быть использовано для под-держки принятии решений, направленных на развитие рынка недвижимости. Сделан вывод о том, что такие модели можно встраивать в системы мониторинга и обновления кадастровых таблиц не-движимости регионов в режиме реального времени, когда изменение экономической ситуации в ре-гионе, стране и мире влечет автоматическое обновление кадастровых данных.Ключевые слова: интеллектуальный анализ данных, аналитика, компьютерное модели-рование, business intelligence, нейронные сети, анализ рынка, оценка. A.O. Alekseev, V.A. Kharitonov, V.L. IasnitskiiPerm National Research Polytechnic University, Perm, Russian Federation DISCUSSION OF DATA MINING, MASS APPRAISAL AND MANAGEMENT OF REAL ESTATE REGIONAL MARKETDiscusses the advantages of using "information approach" to modeling the regional real estate market. Shows the importance of considering the influence of mega-, macro-and meso-economic factors when building computer models of market, suitable for analysis and forecasting of the market value of residential properties, and mass cadastral valuation, the underlying effective tax system to more fairly distribute the tax burden. In addition, mining models "what if" allows you to see how changes in the key pricing factors will affect the market value of objects, what is the decision support aimed at the development of the real estate market. It is concluded that such models can be embedded in a monitoring system of the housing market and updates inventory tables real estate regions in real time, when change of the economic situation in the region, the country and the world entails an automatic update of inventory data.

show abstract

Analyzing Massive Data Sets: An Adaptive Fuzzy Neural Approach for Prediction, with a Real Estate Illustration

Cited by 32 publications

References 22 publications

The spatially conscious machine learning model

The spatially conscious machine learning model

Identifying Real Estate Opportunities Using Machine Learning

Discussion of Data Mining, Mass Appraisal and Management of Real Estate Regional Market

Contact Info

Product

Resources

About