Systematic investigation of keywords selection and processing strategy on search engine forecasting: a case of tourist volume in Beijing

The timeliness, precision, and low cost of search data have great potential for projecting tourist volume. Obtaining valuable information decision-making, particularly predicting, is hampered by the vast amount data. A systematic investigation keyword selection processing has been conducted. Using Beijing volume as an example, 11 different feature extraction algorithms were selected combined with long short-term memory (LSTM), random forest (RF) fuzzy time series (FTS) forecasting total 1612 keywords retrieved from Baidu Index demand mapping using direct word method, range method empirical method. remaining 813 subjected to extraction. Based on results medium (1-day, 7-days 10-days), Kernel principal component analysis (KPCA) locally linear embedding (LLE) are relatively stable when dimensionality reduced 5 dimensions. t-stochastic neighbor (t-SNE), isometric (IsoMap) (LLE), locality preserving projections (LPP), independent correlation (ICA) 10 Accurately many factors (transportation, attraction, food, lodging, travel, tips, tickets, weather) provides a solid foundation tourism optimization scientific management resource tourists' holistic vacation planning.