Project Overview

This project is an AI-powered accommodation monitoring and operations system built to automatically collect Airbnb listing data and analyze price changes, guest review sentiment, and response direction in an integrated workflow.

Previously, operators had to manually search listings by area, open each detail page one by one, review pricing and property information, read guest reviews, interpret review tone individually, and draft response messages by hand.

To improve this process, the project restructured the workflow into a pipeline covering search result collection, detail page crawling, price and listing data structuring, review collection, AI-based review sentiment analysis, AI-generated response drafts, and competitive pricing analysis. This reduced repetitive monitoring work while creating a data foundation that could support both accommodation operations and market response.

Beyond simple page scraping, the system was extended to include review sentiment analysis, response draft generation, and AI-based competitive price estimation, making it more than a data collection tool and turning it into an AX-oriented analysis system that supports operator judgment and action.

Challenges Addressed

Accommodation operations involve more than checking prices. They require understanding competitor trends, interpreting guest reviews, drafting response messages, and evaluating pricing strategy as part of a broader operational workflow.

Previously, operators had to manually search properties in specific areas and repeatedly review listing details and guest feedback. This was time-consuming, and both review interpretation and response writing often relied heavily on individual experience. In addition, price data and review data were managed separately, which made it difficult to review them together and use them effectively for operational decision-making.

For this reason, the project aimed not simply to automate crawling, but to build an operational analysis framework that continuously collects and analyzes accommodation price and review data while supporting real decision-making and response workflows.

Expected Impact and Business Value

Reduced manual monitoring workload
By automating the repeated process of searching listings, checking detail pages, and reviewing guest feedback, the system helps reduce repetitive work and significantly shortens monitoring time.

Operational use of review data
By transforming reviews from simple reference material into data that can support sentiment analysis and response strategy, the system enables a more structured understanding of guest reactions.

Improved speed and quality of guest responses
With AI-generated response drafts, operators can reduce review handling time while adapting responses to their own tone and operational direction.

More advanced pricing decisions
By going beyond basic price collection and applying AI-based competitive pricing analysis, the system helps operators review pricing adjustments and market response strategies on a more data-driven basis.