Applied sentiment analysis on a real estate advertisement recommendation model

ABSTRACT

Recently, the data generated are exploding in the information age. In the post-COVID-19 era, some real estate contracts have been signed online, and online advertisement recommendation has become a new way to reduce the searching cost. Therefore, the model in which real estate online recommendations can be made suitable without user preferences has become a tricky problem. This study uses sentiment and economic data to predict real estate sales and then made an advertisement recommendation from the forecast results. The 2SA-RERec (Two Sentiment Analysis of Real Estate Recommendation) model is proposed, which shows the highest accuracy among the others.

KEYWORDS:

• Real estate online advertisement recommendation model
• sentiment analysis
• seasonal and trend decomposition using loess
• grey system theory
• back-propagation neural network

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Shenzhen Government [20200829144221001, GXWD20201230155427003 – 20200829144221001, HA11409015].