Code package for supplementary materials

This package contains the main scripts used to support the analytical workflow reported in the manuscript:

1. Code_1_BERTopic_topic_modeling_pipeline_revised.py
   BERTopic pipeline based on Sentence-BERT, UMAP, and HDBSCAN.

2. Code_2_Review_preprocessing_and_sentence_segmentation_revised.py
   Review cleaning, removal of platform-generated default positive comments and other low-information entries,
   de-duplication, and rule-based sentence segmentation.

3. Code_3_Topic_sentiment_scoring_revised.py
   Topic assignment merge and sentiment scoring at the review-segment level.

4. Code_4_Destination_level_variable_construction_revised.py
   Construction of topic-level condition scores and the destination-level EQ index.

5. Code_5_Fuzzy_set_calibration_revised.py
   Fuzzy-set calibration based on the 25th percentile, median, and 75th percentile.

6. Code_6_NCA_analysis_revised.py
   Necessary Condition Analysis using the CE-FDH approach.

7. Code_7_fsQCA_analysis_and_robustness.R
   fsQCA truth tables, necessity analysis, complex/parsimonious solutions, and robustness checks.

Suggested software environment
Python:
- pandas
- numpy
- matplotlib
- bertopic
- sentence-transformers
- scikit-learn
- umap-learn
- hdbscan
- transformers
- openpyxl

R:
- QCA
- openxlsx
- dplyr
- stringr