Research on sentiment index and real estate demand forecasting based on BERT-BiLSTM and ADL-MIDAS models

Abstract To investigate the impact of real estate market sentiment on demand forecasting, this paper constructs a Weibo sentiment index incorporating emotional polarity and verifies its predictive advantage for market demand. Based on the Bidirectional Encoder Representations from Transformers - Bid...

Full description

Saved in:
Bibliographic Details
Main Authors: Mengkai Chen, Jun Wang, Feilong Zhao, Gaopeng Jiang
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-16153-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849226512225533952
author Mengkai Chen
Jun Wang
Feilong Zhao
Gaopeng Jiang
author_facet Mengkai Chen
Jun Wang
Feilong Zhao
Gaopeng Jiang
author_sort Mengkai Chen
collection DOAJ
description Abstract To investigate the impact of real estate market sentiment on demand forecasting, this paper constructs a Weibo sentiment index incorporating emotional polarity and verifies its predictive advantage for market demand. Based on the Bidirectional Encoder Representations from Transformers - Bidirectional Long Short-Term Memory (BERT-BiLSTM) and the characteristics of China’s housing market, we classify sentiments in crawled Weibo texts and train a sentiment analysis model specifically for the Chinese real estate domain. This model accurately extracts positive, neutral, and negative sentiment features to build a high-frequency sentiment index. Simultaneously, Internet Concern index is constructed using Baidu search data as a non-directional sentiment proxy variable. Further adopting the Autoregressive Distributed Lag Mixed Data Sampling model (ADL-MIDAS), we compare the predictive performance of these two sentiment indices alongside macroeconomic variables on market demand. Experimental results show that: (1) The BERT-BiLSTM model achieves 78.5% accuracy in sentiment classification, with its F1-score outperforming traditional methods (SVM, LSTM, etc.) by over 30%; (2) The Weibo sentiment index yields a Root Mean Squared Forecast Error (RMSFE) of 1.6%-4.7% under the ADL-MIDAS framework, significantly lower than the Internet Concern index (6.7%-7.0%). The study demonstrates that integrating deep learning with high-frequency social media sentiment indiex containing emotional polarity can more effectively capture market expectation fluctuations, while simultaneously yielding superior performance for real estate market demand forecasting.
format Article
id doaj-art-99d659959d0a4746891a3ff554d52780
institution Kabale University
issn 2045-2322
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-99d659959d0a4746891a3ff554d527802025-08-24T11:17:33ZengNature PortfolioScientific Reports2045-23222025-08-0115111610.1038/s41598-025-16153-8Research on sentiment index and real estate demand forecasting based on BERT-BiLSTM and ADL-MIDAS modelsMengkai Chen0Jun Wang1Feilong Zhao2Gaopeng Jiang3School of Management Science and Engineering, Anhui University of TechnologySchool of Management Science and Engineering, Anhui University of TechnologySchool of Management Science and Engineering, Anhui University of TechnologySchool of Management Science and Engineering, Anhui University of TechnologyAbstract To investigate the impact of real estate market sentiment on demand forecasting, this paper constructs a Weibo sentiment index incorporating emotional polarity and verifies its predictive advantage for market demand. Based on the Bidirectional Encoder Representations from Transformers - Bidirectional Long Short-Term Memory (BERT-BiLSTM) and the characteristics of China’s housing market, we classify sentiments in crawled Weibo texts and train a sentiment analysis model specifically for the Chinese real estate domain. This model accurately extracts positive, neutral, and negative sentiment features to build a high-frequency sentiment index. Simultaneously, Internet Concern index is constructed using Baidu search data as a non-directional sentiment proxy variable. Further adopting the Autoregressive Distributed Lag Mixed Data Sampling model (ADL-MIDAS), we compare the predictive performance of these two sentiment indices alongside macroeconomic variables on market demand. Experimental results show that: (1) The BERT-BiLSTM model achieves 78.5% accuracy in sentiment classification, with its F1-score outperforming traditional methods (SVM, LSTM, etc.) by over 30%; (2) The Weibo sentiment index yields a Root Mean Squared Forecast Error (RMSFE) of 1.6%-4.7% under the ADL-MIDAS framework, significantly lower than the Internet Concern index (6.7%-7.0%). The study demonstrates that integrating deep learning with high-frequency social media sentiment indiex containing emotional polarity can more effectively capture market expectation fluctuations, while simultaneously yielding superior performance for real estate market demand forecasting.https://doi.org/10.1038/s41598-025-16153-8BERT-BiLSTMSentiment classificationSentiment polarityADL-MIDAS
spellingShingle Mengkai Chen
Jun Wang
Feilong Zhao
Gaopeng Jiang
Research on sentiment index and real estate demand forecasting based on BERT-BiLSTM and ADL-MIDAS models
Scientific Reports
BERT-BiLSTM
Sentiment classification
Sentiment polarity
ADL-MIDAS
title Research on sentiment index and real estate demand forecasting based on BERT-BiLSTM and ADL-MIDAS models
title_full Research on sentiment index and real estate demand forecasting based on BERT-BiLSTM and ADL-MIDAS models
title_fullStr Research on sentiment index and real estate demand forecasting based on BERT-BiLSTM and ADL-MIDAS models
title_full_unstemmed Research on sentiment index and real estate demand forecasting based on BERT-BiLSTM and ADL-MIDAS models
title_short Research on sentiment index and real estate demand forecasting based on BERT-BiLSTM and ADL-MIDAS models
title_sort research on sentiment index and real estate demand forecasting based on bert bilstm and adl midas models
topic BERT-BiLSTM
Sentiment classification
Sentiment polarity
ADL-MIDAS
url https://doi.org/10.1038/s41598-025-16153-8
work_keys_str_mv AT mengkaichen researchonsentimentindexandrealestatedemandforecastingbasedonbertbilstmandadlmidasmodels
AT junwang researchonsentimentindexandrealestatedemandforecastingbasedonbertbilstmandadlmidasmodels
AT feilongzhao researchonsentimentindexandrealestatedemandforecastingbasedonbertbilstmandadlmidasmodels
AT gaopengjiang researchonsentimentindexandrealestatedemandforecastingbasedonbertbilstmandadlmidasmodels