Airbnb Price Prediction Using Machine Learning and Sentiment Analysis

Kalehbasti, Pouya Rezazadeh; Nikolenko, Liubov; Rezaei, Hoormazd

doi:10.1007/978-3-030-84060-0_11

Cited by 23 publications

(14 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When in addition to the spatial data we take into account high-cardinality features such as a car's model, in a single covariance structure, the improvement in test MSE is substantial. The mean test MSE achieved for the Airbnb dataset is far better than the best test MSE (0.147) reported by Kalehbasti et al (2019), who also tried using boosting and support vector machines. More details such as mean running times appear in Table 28 in Appendix 3, and plots of the distribution of n j measurements in location and predicted y te versus true appear in Figure 9 in Appendix 5.…”

Section: Spatial Data and Spatial-categorical Combinationsmentioning

confidence: 55%

Integrating Random Effects in Deep Neural Networks

Simchoni¹

2022

Preprint

View full text Add to dashboard Cite

Modern approaches to supervised learning like deep neural networks (DNNs) typically implicitly assume that observed responses are statistically independent. In contrast, correlated data are prevalent in real-life large-scale applications, with typical sources of correlation including spatial, temporal and clustering structures. These correlations are either ignored by DNNs, or ad-hoc solutions are developed for specific use cases. We propose to use the mixed models framework to handle correlated data in DNNs. By treating the effects underlying the correlation structure as random effects, mixed models are able to avoid overfitted parameter estimates and ultimately yield better predictive performance. The key to combining mixed models and DNNs is using the Gaussian negative log-likelihood (NLL) as a natural loss function that is minimized with DNN machinery including stochastic gradient descent (SGD). Since NLL does not decompose like standard DNN loss functions, the use of SGD with NLL presents some theoretical and implementation challenges, which we address. Our approach which we call LMMNN is demonstrated to improve performance over natural competitors in various correlation scenarios on diverse simulated and real datasets. Our focus is on a regression setting and tabular datasets, but we also show some results for classification. Our code is available at https://github.com/gsimchoni/lmmnn.

show abstract

Section: Spatial Data and Spatial-categorical Combinationsmentioning

confidence: 55%

Integrating Random Effects in Deep Neural Networks

Simchoni¹

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…These studies focus on testing research hypotheses for evaluating the influence of respective features without focusing on the predictive modeling of the listing prices. On the other hand, authors have predicted Airbnb listing prices using amenities-driven features (Kalehbasti et al , 2021; Islam et al , 2022). Thus, no focus has been given to Airbnb listing price modeling without using amenity-driven features.…”

Section: Introductionmentioning

confidence: 99%

An ensemble machine learning framework for Airbnb rental price modeling without using amenity-driven features

Ghosh

Jana

Abedin

2023

IJCHM

View full text Add to dashboard Cite

Purpose The prediction of Airbnb listing prices predominantly uses a set of amenity-driven features. Choosing an appropriate set of features from thousands of available amenity-driven features makes the prediction task difficult. This paper aims to propose a scalable, robust framework to predict listing prices of Airbnb units without using amenity-driven features. Design/methodology/approach The authors propose an artificial intelligence (AI)-based framework to predict Airbnb listing prices. The authors consider 75 thousand Airbnb listings from the five US cities with more than 1.9 million observations. The proposed framework integrates (i) feature screening, (ii) stacking that combines gradient boosting, bagging, random forest, (iii) particle swarm optimization and (iv) explainable AI to accomplish the research objective. Findings The key findings have three aspects – prediction accuracy, homogeneity and identification of best and least predictable cities. The proposed framework yields predictions of supreme precision. The predictability of listing prices varies significantly across cities. The listing prices are the best predictable for Boston and the least predictable for Chicago. Practical implications The framework and findings of the research can be leveraged by the hosts to determine rental prices and augment the service offerings by emphasizing key features, respectively. Originality/value Although individual components are known, the way they have been integrated into the proposed framework to derive a high-quality forecast of Airbnb listing prices is unique. It is scalable. The Airbnb listing price modeling literature rarely witnesses such a framework.

show abstract

“…There are many studies using various machine learning techniques in price estimation applications. Home price prediction [7][8][9][10], stock price prediction [11][12][13][14], stock market prediction [15], bitcoin price prediction [16], price estimate for vacation rentals [17], car price prediction [18][19] are some of them. In this study the main focus is on used vehicle price estimation.…”

Section: Introductionmentioning

confidence: 99%

“…In this study, LR, tree-based models, Decision Support Machines and ANN methods were applied to get the best results in terms of Mean Square Line, Mean Absolute Error (MAE) and R2 score. Among the methods tested, SVR gave the best result with an R2 score of 69% [17]. In a study conducted for vehicle price estimation, some machine learning algorithms are used and compared for an application that finds the best price that a truck company can give when buying used vehicles from customers.…”

Section: Introductionmentioning

confidence: 99%

Using Linear Regression For Used Car Price Prediction

MUTİ

Yildiz

2023

International Journal of Computational and Experimental Science and Engineering

View full text Add to dashboard Cite

Recently, there have been studies on the use of machine learning algorithms for price prediction in many different areas such as stock market, rent a house and used car sales. Studies give information about which algorithm is more successful in price prediction using different machine learning methods. The most commonly used method for price prediction is the linear regression model. In our study, we examined the effectiveness of the linear regression model for used car price prediction. In our study, we applied the linear regression model on a data set that includes the features and price information of vehicles in Turkey for the year 2020. As a result, when we selected 1/3 of the data set as the test data, we observed that the R2 score for the prediction success of our model was 73%. More successful results can be obtained with different data sets or a more detailed data preprocessing.

show abstract

Airbnb Price Prediction Using Machine Learning and Sentiment Analysis

Cited by 23 publications

References 16 publications

Integrating Random Effects in Deep Neural Networks

Integrating Random Effects in Deep Neural Networks

An ensemble machine learning framework for Airbnb rental price modeling without using amenity-driven features

Using Linear Regression For Used Car Price Prediction

Contact Info

Product

Resources

About