Deep Learning in Predicting Real Estate Property Prices: A Comparative Study

2023 | HICSS | Citations: 0

Authors: Shi, D.; Zhang, H.; Guan, J.; Zurada, J.; Chen, Z.; Li, X.

Abstract: The dominant methods for real estate property price prediction or valuation are ... Expand

Abstract: The dominant methods for real estate property price prediction or valuation are multi-regression based. Regression-based methods are, however, imperfect because they suffer from issues such as multicollinearity and heteroscedasticity. Recent years have witnessed the use of machine learning methods but the results are mixed. This paper introduces the application of a new approach using deep learning models to real estate property price prediction. The paper uses a deep learning approach for modeling to improve the accuracy of real estate property price prediction with data representing sales transactions in a large metropolitan area. Three deep learning models, LSTM, GRU and Transformer, are created and compared with other machine learning and traditional models. The results obtained for the data set with all features clearly show that the RF and Transformer models outperformed the other models. LSTM and GRU models produced the worst results, suggesting that they are perhaps not suitable to predict the real estate price. Furthermore, the implementations of Transformer and RF on a data set with feature reduction produced even more accurate prediction results. In conclusion, our research shows that the performance of the Transformer model is close to the RF model. Both models produce significantly better prediction results than existing approaches in terms of accuracy. Collapse

Topics: Python open source

Methods: deep learning transformer model machine learning autoregressive moving average model experiment

Trading on Cryptocurrency Markets: Analyzing the Behavior of Bitcoin Investors

2019 | International Conference on Information Systems | Citations: 1

Authors: Keller, Alexander; Scholz, Michael

Abstract: Driven by innovative information technologies, the financial industry is facing ... Expand

Abstract: Driven by innovative information technologies, the financial industry is facing a recent disruptive fintech revolution. One emerging technology within this field is cryptocurrency, aiming to change the future means of payment. In this paper, we study Bitcoin exchange trading and examine what factors influence the behavior of different cryptocurrency investor types. To answer this question, market bids are considered in form of investors' offers and orders as a proxy for their trading behavior. First, an unsupervised clustering technique is applied in order to group different types of investors based on similarities in trading behavior. Second, a supervised classification mechanism is used on social media news to measure the sentiment influencing trading decisions. Among other indicators this bullishness is integrated in an autoregressive distributed lag (ARDL) model to identify the factors influencing the trading behavior of investor types. Besides large investors, foreign traders and speculators, cryptocurrency-specific market participants are characterized in the form of miners. With identifying indicators driving investors' actions (i.e., macro-financial fundamentals, technical trading indicators, technological measures and market sentiment), this study contributes to recent research by explaining the trading behavior on cryptocurrency markets and its impact on exchange rates. Collapse

Topics: Bitcoin Twitter cryptocurrency application programming interface social media

Methods: econometric modeling cluster analysis simulation econometric model longitudinal research

An empirical study on predicting cloud incidents

2019 | International Journal of Information Management | Citations: 1

Authors: Roumani, Yaman; Nwankpa, Joseph K.

Abstract: With the increasing rate of adoption and growth of cloud computing services, bus ... Expand

Abstract: With the increasing rate of adoption and growth of cloud computing services, businesses have been shifting their information technology (IT) infrastructure to the cloud. Although cloud vendors promise high availability and reliability of their cloud services, cloud-related incidents involving outages and service disruptions remain a challenge. Understanding cloud incidents and the ability to predict them would be helpful in deciding how to manage and circumvent future incidents. In this study, we propose a hybrid model that employs machine learning and time series methods to forecast cloud incidents. We evaluate the proposed model using a sample of 2261 incidents collected from two cloud providers namely, Netflix and Hulu. Unique to this study is that our model relies solely on historical data that is independent of the underlying cloud infrastructure. Results suggest that the proposed hybrid model outperforms individual forecasting models: neural network, time series and random forest. Results also reveal important temporal insights from the proposed model and highlights the practical relevance of historical data to forecast and manage cloud incidents. Collapse

Topics: cloud computing digital service Hulu Netflix IT manager

Methods: autoregressive integrated moving average longitudinal research time series analysis artificial neural network machine learning

Oil Prices Forecasting: A Comparative Analysis

2018 | International Conference on Information Systems | Citations: 0

Authors: Tripathi, Manas; Kaur, Inderjeet

Abstract: Oil prices have impacts on both economic and non-economic factors. Therefore, th ... Expand

Abstract: Oil prices have impacts on both economic and non-economic factors. Therefore, the accurate prediction of these prices is an important matter. This study forecasts the crude oil prices by using four different time series models such as artificial neural network (NN), exponential smoothing, ARMA-GARCH, and random walk. The forecasts have also been generated for an ensemble model which is the combination of these four models. The data sample of the study involves daily spot price observations from January 2013 to January 2018. After training, the predictive power of the models has been tested for different durations of the forecasting period. The results suggest that the ensemble model performs better for the majority of the cases. This confirms the better predicting power of nonlinear models for oil prices returns. The study aims to contribute to the literature on oil price forecasting. The better oil price predictions have implications for the oil-dependent industries, government policies, and investor policies. Collapse

Topics: information management

Methods: longitudinal research autoregressive moving average model time series analysis generalized autoregressive conditional heteroskedasticity model artificial neural network

Predicting stock market returns from malicious attacks: A comparative analysis of vector autoregression and time-delayed neural networks

2011 | Decision Support Systems | Citations: 0

Authors: Khansa, Lara; Liginlal, Divakaran

Abstract: With the growing importance of Internet-based businesses, malicious code attacks ... Expand

Abstract: With the growing importance of Internet-based businesses, malicious code attacks on information technology infrastructures have been on the rise. Prior studies have indicated that these malicious attacks are associated with detrimental economic effects on the attacked firms. On the other hand, we conjecture that more intense malicious attacks boost the stock price of information security firms. Furthermore, we use artificial neural networks and vector autoregression analyses as complementary methods to study the relationship between the stock market returns of information security firms and the intensity of malicious attacks, computed as the product of the number of malicious attacks and their severity levels. A major contribution of this work is the resulting time-delayed artificial neural network model that allows stock return predictions and is particularly useful as an investment decision support system for hedge funds and other investors, whose portfolios are at risk of losing market value during malicious attacks. Collapse

Topics: data security computer virus protection intrusion detection system email spam IT security threat

Methods: longitudinal research time series analysis machine learning artificial neural network computational algorithm

Theories: deterrence theory

Outcome Based Systems Evaluation to Assess Information Technology Impact Using Arima Methods

2003 | Communications of the Association for Information Systems | Citations: 0

Authors: Evans, Gerald E.; Costa, Bruce A.

Abstract: A new method of system evaluation that focuses on the impact the system has on a ... Expand

Abstract: A new method of system evaluation that focuses on the impact the system has on a data series that served as the rationale for systems implementation was designed and modeled by the authors. Called outcome-based evaluation, this method is founded on the concept of intervention analysis and employs interrupted time series designs to determine the impact of an information system on specific organizational goals. Based on a review of the literature on the evolution of systems evaluation methods from focusing on user goals to user satisfaction and system usage, we conclude that user satisfaction and system usage are necessary but not sufficient criteria to establish system effectiveness or success. Thus, we establish a need for the proposed new method of system evaluation. Collapse

Topics: inventory management system job satisfaction point of sale system information system use user satisfaction

Methods: autoregressive integrated moving average longitudinal research time series analysis case study survey

How Software Evolves: An Exploratory Analysis of Software Change Histories

1996 | Americas Conference on Information Systems | Citations: 0

Authors: Slaughter, Sandra; Sim, Weelin

Abstract: ... Expand

Abstract: Collapse

Topics: system maintenance IS management software developer

Methods: longitudinal research mathematical model autoregressive integrated moving average qualitative coding partial autocorrelation function