CN115577190A - Tourist behavior data extraction method - Google Patents

Abstract

The application belongs to the technical field of tourism data processing, and discloses a method for extracting tourist behavior data. The method includes: obtaining the check-in data of tourist attractions, performing structural processing, and obtaining a check-in space-time database; acquiring the first travel note sample from a travel website, marking the time information and location information of the travel note text in it, and obtaining the marked travel space-time path, based on the mark method, forming a preliminary analysis module; obtaining the second travel note sample, running the preliminary analysis module to obtain the analytical travel space-time path of the second travel note sample, improving the preliminary analysis module based on the analysis of travel space-time path, and obtaining the final analysis module; applying the final analysis module In the travel notes samples of the preset time window and the preset destination range, a travel notes spatio-temporal database is obtained; based on the check-in spatio-temporal database and the travel notes spatio-temporal database, a visualized tourist spatio-temporal behavior path map is obtained. Provide structured data for subsequent patent analysis in the field of tourism.

Description

A Method for Extracting Tourist Behavior Data

technical field

This application relates to the technical field of tourism data processing, in particular to a method for extracting tourist behavior data.

Background technique

In recent years, with the rapid development of the economy and transportation, the willingness of domestic tourists to travel has increased significantly, the number of tourists has also continued to increase, and the income of tourism-related industries has also increased. At the same time, with the development of Internet technology, tourists leave a large amount of tourism-related data on the Internet during their travels. These tourism-related data can be used for analysis and research such as tourism marketing planning, tourist number prediction, route planning, and scenic spot evaluation. To provide tourists with better tourism services and develop better tourism products. However, although the information obtained by the existing technology has a wide range of information sources, the data collection focuses on the extraction of static tourism information, and lacks the structured processing of dynamic tourism information between different scenic spots.

Contents of the invention

To this end, the embodiment of the present application provides a method for extracting tourist behavior data, which realizes structured extraction and visualization processing of dynamic tourist information.

In the first aspect, the present application provides a method for extracting tourist behavior data.

The application is achieved through the following technical solutions:

A method for extracting tourist behavior data, said method comprising:

Obtain the check-in data of tourist attractions, and carry out structured processing to the check-in data of the tourist attractions, and obtain the check-in space-time database based on the check-in data of the tourist attractions;

Obtain the first travel note sample from a travel website, mark the time information and location information of each travel note text in the first travel note sample, and obtain the marked travel space-time path, based on the marking method of the travel space-time path, form a preliminary Analysis module;

Obtaining a second travel note sample, running the preliminary analysis module to obtain the analytical travel space-time paths of all travel texts of the second travel note sample, improving the preliminary analysis module based on the analysis of travel space-time paths, and obtaining the final analysis module;

Applying the final parsing module to travel notes samples in a preset time window and a preset destination range to obtain a travel notes spatio-temporal database based on travel notes;

A tourist flow behavior database is constructed based on the check-in spatio-temporal database and the travel notes spatio-temporal database, and a visualized tourist spatio-temporal behavior path map is obtained based on the tourist flow behavior database.

In a preferred example of the present application, it may be further configured that the tourist flow behavior database is constructed based on the check-in space-time database and the travel note space-time database, and the visualized tourist space-time behavior path map is obtained based on the tourist flow behavior database. Before step, also include:

Collect comment data of tourist attractions, calculate the proportion of comment data of a single tourist attraction in all tourist attractions in the local city, and obtain the benchmark number of tourists visiting the tourist attraction based on the proportion of comment data;

Based on the check-in spatio-temporal database, the first tourist visitor number is obtained, based on the travel notes spatio-temporal database, the second tourist visitor number is obtained, and the difference between the first tourist visitor number and the second tourist visitor number and the reference tourist visitor number is calculated respectively. The deviation ratio of the number of visitors.

In a preferred example of the present application, it may be further set that, after calculating the deviation ratios between the first number of tourist visitors and the second number of tourist visitors and the reference number of tourist visitors, it may further include:

Comparing the deviation ratio with the preset deviation ratio, if the deviation ratio exceeds the preset deviation ratio, the check-in spatio-temporal database and the travel notes spatio-temporal database need to be further improved;

If the deviation ratio is within the preset deviation ratio, a visitor flow behavior database is obtained based on the check-in spatio-temporal database and the travel notes spatio-temporal database.

In a preferred example of the present application, it may be further set as the step of obtaining the check-in data of the tourist attraction, and performing structural processing on the check-in data of the tourist attraction, and obtaining the check-in spatio-temporal database based on the check-in data of the tourist attraction, include:

According to the name of the scenic spot in the target area, the scenic spot check-in ID of each scenic spot is obtained, and the list of scenic spot, scenic spot sign-in ID and scenic spot number is obtained;

Obtain the check-in data of the corresponding user IDs of all scenic spots check-in IDs in the scenic spot in the time window, and obtain the initial user database of the scenic spot, which includes user ID, check-in time, check-in location and check-in content in the initial user database;

Obtain the personal information of all user IDs in the initial user database, supplement the initial user database as a supplementary table, and obtain the check-in spatio-temporal database of tourist attraction check-in data.

In a preferred example of the present application, it can be further set that the step of obtaining the ID name of the scenic spot of each scenic spot also includes:

Establish the main sign-in ID and several subordinate sign-in IDs of the scenic spot, and summarize several of the subordinate sign-in IDs into the main sign-in ID, and use the main sign-in ID as the scenic spot sign-in ID of each scenic spot.

In a preferred example of the present application, it may be further set that the step of obtaining the check-in data of the user IDs corresponding to the check-in IDs of all the scenic spots in the scenic spot within the time window further includes:

If the same user ID corresponds to multiple different scenic spot check-in IDs, then the user ID is associated with the multiple different scenic spot check-in IDs;

If the same user ID corresponds to multiple identical scenic spot check-in IDs, the multiple identical scenic spot check-in IDs are deduplicated.

In a preferred example of the present application, it may be further configured to obtain a first travel note sample from a travel website, mark the time information and location information of each travel note text in the first travel note sample, and obtain a marked travel space-time path , based on the marking method of marking the travel space-time path, the step of forming a preliminary parsing module includes:

Mark the time keywords of each piece of travel text in the first travel note sample based on precise date, precise time, fuzzy time and relative time, place the time keyword in the time lexicon, and use precise date and precise Time is used as the segmentation point to segment the travel note text into text segments;

identifying location keywords for the text segment based on precise locations, vague locations, and associated locations, and placing the location keywords in a location thesaurus;

Based on the extraction method of the time keyword and location keyword, a preliminary parsing module is constructed.

In a preferred example of the present application, it can be further set as, the acquisition of the second travel note sample, and sequentially running the preliminary parsing module to obtain the analytical travel time and space paths of all travel texts of the second travel note sample, based on the analysis of travel time and space The path is to improve the preliminary parsing module, and the steps of obtaining the final parsing module include:

If the second travel note sample includes travel photos, extract time information and latitude and longitude information in the travel photos, arrange the travel photos based on time order, and generate image spatiotemporal paths of the travel photos.

In a preferred example of the present application, it may be further set that the step of improving the preliminary analysis module based on analyzing the travel space-time path, and obtaining the final analysis module includes:

Extract part of the travel text from the second sample of travel notes and mark it to obtain the marked travel space-time path, and compare the analyzed travel time-space path obtained by the preliminary analysis module with the marked travel time-space path, and based on the comparison result, the preliminary analysis module Carry out improvement to obtain the final time thesaurus and the final parsing module.

In a preferred example of the present application, it may be further set that, based on the check-in spatio-temporal database and the travel notes spatio-temporal database, the step of obtaining a visualized tourist spatio-temporal behavior path map includes:

Based on the spatio-temporal database of check-ins and the spatio-temporal database of travel notes, the path map of single or multiple tourists is formed through ArcGIS software tools, and the time axis and map base map are added to the path map to form a visualized path map of tourist spatio-temporal behavior .

In summary, compared with the prior art, the beneficial effects brought by the technical solutions provided by the embodiments of the present application at least include:

Obtain the check-in data of tourist attractions, carry out structured processing, and obtain the check-in spatio-temporal database; mark the time information and location information of the travel text obtained from the travel website, form an analysis module based on the marking method, and then use the analysis module to analyze the travel sample, Obtain the spatio-temporal database of travel notes; form the tourist flow behavior database based on the check-in spatio-temporal database and the travel notes spatio-temporal database, and perform spatial analysis and visualization based on the tourist flow behavior database. On the basis of integrating objective spatio-temporal geographic information, based on individual dynamic tourism information data among different scenic spots, the extraction method of subjective check-in data and travel data content is structured, which complements the lack of individual data collection methods in other data collection methods. The focus on subjective information realizes the refinement and concreteness of the space-time path.

Description of drawings

Fig. 1 is a schematic flow chart of a method for extracting tourist behavior data provided by an exemplary embodiment of the present application;

FIG. 2 is a schematic flow diagram of establishing a check-in spatio-temporal database provided by an exemplary embodiment of the present application;

Fig. 3 is a schematic flow chart of establishing a spatio-temporal database of travel notes provided by an exemplary embodiment of the present application.

detailed description

This specific embodiment is only an explanation of this application, and it is not a limitation of this application. Those skilled in the art can make modifications to this embodiment without creative contribution according to needs after reading this specification, but as long as the rights of this application All claims are protected by patent law.

In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments It is a part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.

In addition, the term "and/or" in this application is only an association relationship describing associated objects, indicating that there may be three relationships, for example, A and/or B may indicate: A exists alone, and A and B exist simultaneously , there are three cases of B alone. In addition, the character "/" in this application, unless otherwise specified, generally indicates that the contextual objects are an "or" relationship.

In this application, the terms "first" and "second" are used to distinguish the same or similar items with basically the same function and function. It should be understood that "first", "second" and "nth" There are no logical or timing dependencies, nor are there restrictions on quantity or order of execution.

The embodiments of the present application will be further described in detail below in conjunction with the accompanying drawings.

In one embodiment of the present application, a method for extracting tourist behavior data is provided, as shown in Figure 1, the main steps are described as follows:

S10: Obtain the check-in data of the tourist attraction, and perform structural processing on the check-in data of the tourist attraction, to obtain a check-in space-time database based on the check-in data of the tourist attraction.

Specifically, take the acquisition of tourist attraction check-in data from Weibo as an example for illustration. Use Python to obtain the check-in data of tourist attractions in the target area, and then perform structured processing on the check-in data of tourist attractions, as shown in Figure 1. The specific steps are:

According to the names of the scenic spots in the target area, search the names of the scenic spots one by one on the Weibo sign-in page, obtain the scenic spot sign-in ID and scenic spot number corresponding to each scenic spot name, create a list, and put the scenic spot name and scenic spot sign-in ID and scenic spot number Associating, as a row element of the list, a list containing the name of the scenic spot, the check-in ID of the scenic spot and the number of the scenic spot is obtained;

Obtain the user sign-in data corresponding to all scenic spot sign-in IDs of all scenic spots in the time window, including the user ID, sign-in time, sign-in location and sign-in content related to the sign-in ID of the scenic spot, and obtain the initial user database of the scenic spot. What needs to be explained is that social media data represented by Weibo can only be crawled in real time, so it is necessary to set a time window for data collection. Generally, according to the purpose of data analysis, it can be divided into three types: Year Unit, set the time of a certain year; month unit, set the time of a certain month in a certain year; holiday unit, set the time of 3-7 days of a certain legal holiday. Further obtain the personal information corresponding to all user IDs in the initial user database, including user gender, user origin, date of birth, and graduate school information, and supplement the initial user database with personal information as an attached table, and obtain Check-in spatio-temporal database of check-in data.

Preferably, the main sign-in ID and several subordinate sign-in IDs of the scenic spot are established, and the several subordinate sign-in IDs are summarized into the main sign-in ID, and the main sign-in ID is used as the scenic spot sign-in ID of each scenic spot.

Specifically, since there are multiple place names for the same place on the Weibo check-in page, or a certain scenic spot contains multiple small scenic spots, there are multiple scenic spot check-in IDs in some scenic spots. It is possible to establish the main sign-in ID belonging to the first level and the subordinate sign-in ID belonging to the second level of the scenic spot. When several subordinate sign-in IDs belong to the same main sign-in ID, they will be aggregated together into the main sign-in ID, and the main sign-in ID will be used as each scenic spot. The scenic area sign-in ID.

Preferably, if the same user ID corresponds to multiple different scenic spot check-in IDs, then the user ID is associated with multiple different scenic spot check-in IDs; if the same user ID corresponds to multiple identical scenic spot check-in IDs, then the multiple The same scenic spot check-in ID is deduplicated. Specifically, you can use python to traverse the data, filter repetitive data, and deduplicate processing can avoid saving duplicate data in the database and causing a large amount of redundant data.

Because the same user may use Weibo to check-in in multiple scenic spots in the target area, there are multiple different scenic spot check-in IDs corresponding to the same user ID in the initial user database, and multiple different scenic spot check-in IDs of the same user ID are associated. At the same time, if the same user ID corresponds to multiple identical scenic spot check-in IDs, duplicate scenic spot check-in IDs will be deduplicated to form a behavior track based on individual users' flow in the scenic spot space, and it will be represented in the form of tourism spatio-temporal data tables. The travel time-space data table is divided into rows according to user ID, and arranges "time point 1, location 1, time point 2, location 2..." in sequence, listing the travel time-space information corresponding to each user ID in detail.

As shown in Table 1, it shows part of the data of the check-in spatio-temporal database, which includes information such as "user ID", "user source place", "time point 1", "check-in location 1".

Table 1

Note: ① It should be noted that the user’s source of customers should not include the check-in location completely, otherwise it should be excluded as a non-tourist group; ② Users should pay attention to setting the time limit for deduplication by themselves, and pay attention to the tourists in the last row of the similar table who go on vacation The situation of staying in the area for a long time.

As shown in Figure 2, the travel notes data comes from the text content and image links of authoritative guide websites, including but not limited to the following guide websites: Mafengwo, Ctrip, Qunar, and Qyer.com. The specific way to obtain it is as follows:

S20: Use Python to automatically extract different types of travel notes, randomly select the first sample of travel notes, mark the time information and location information of each travel note text, obtain the marked travel space-time path of the travel note text, and summarize the marked travel space-time The marking method of time information and location information in the route, based on the marking method, forms an initial parsing module.

Specifically, the time keywords and location keywords in the time and location thesaurus should be connected in series corresponding to each user ID in chronological order and location changes. Manually identify the time keywords in each travel text, and mark the time keywords according to "accurate date", "precise time", "fuzzy time" and "relative time", and classify them into corresponding time words library, and use the "precise date" and "precise time" as the segmentation points to divide the travel note text into text segments sorted in chronological order. For example, the expression form of "exact date" can be September 6th, 9/6, the second day and/or DAY2; the expression form of "exact time" can be 10:30, 14:30 and/or or four in the afternoon; "fuzzy time" can be expressed as am, pm, evening, early morning, breakfast, lunch, and/or night view; "relative time" can be expressed as 15 minutes later, about 2 hours away and/or Or play for about 1.5 hours. The relative time can be calculated by the median method to calculate the absolute time. "Precise Date", "Precise Time", "Fuzzy Time" and "Relative Time" are applied as follows: "Precise Date" and "Precise Time" are used as the dividing points, and "Fuzzy Time" can estimate the specific time according to the morning and afternoon. The time needs to use the median method to calculate the absolute time. The usage of the time lexicon is matching and calculation, and the ultimate goal is to obtain a specific time point.

Manually identify the location keywords in the text segment of each travel text, and mark the location keywords in the travel text according to "precise location", "fuzzy location" and "associated location", and place the location keywords to the thesaurus of places. The location lexicon is used in the way of association and comparison, and the ultimate goal is to unify each location to the same level and correspond to time keywords. It needs to be illustrated that the expression form of "precise location" can be Hong Kong (Hong Kong), Fujian (Fujian Province, Fujian), Canton Tower Scenic Area (Xiaomanyao) and/or Huangguoshu Waterfall (Huangguoshu Scenic Area); " can be expressed in the main entrance, end point and/or mountain top; "associated places" can be expressed in the form of arriving at a certain place, going to a certain place, returning to a certain place, visiting a certain place, boarding a certain place, going around a certain place , Live in a certain place, from A to B. The application methods of "precise location", "fuzzy location" and "associated location" are association and comparison, that is, in the descriptive sentence group corresponding to the time, the location at that time is extracted according to the matching situation, forming a space-time path; fuzzy The location needs to supplement the specific location through the context; if there are multiple locations, it needs to further analyze whether it needs to add time or delete the location; if the exact location is in the previous text segment and the current text segment does not have an exact location, it can be compared with the precise location information of the previous paragraph merge.

Based on the extraction method of time keywords and location keywords, write in chronological order, and write the preliminary analysis module for machine learning based on the logic of location changes. The specific steps are:

F1: Set the preview area to filter text segments with spatio-temporal information;

F2: the text query input box of time lexicon is set, automatically identify the time keywords in the text segment with spatio-temporal information;

F3: the output result of the text query input box of the time lexicon is set and the query input interface link matched in the task template, and the time keywords identified are corresponding to the divided time lexicon;

F4: Set the output conversion interface to link with the corresponding conversion program, output the time information corresponding to each user ID entered into the preset table, and output it in the matching order from date to time, forming a descriptive sentence group with time as the core ;

F5: the location thesaurus query input box is set, automatically identify the location keyword in the text segment with spatio-temporal information;

F6: the output result of the location thesaurus text query input box is set and the query input interface link matched in the task template, and the location keywords identified are corresponding to the divided location thesaurus;

F7: set the output conversion interface to link with the corresponding conversion program, and output the place information corresponding to different times of each user ID entered into the preset table to form a space-time path.

S30: Obtain the second travel note sample, run the above preliminary analysis module to obtain the analytical travel space-time path of all travel texts of the second travel note sample, improve the preliminary analysis module based on the analysis of travel space-time path, and obtain the final analysis module.

Specifically, the above-mentioned preliminary analysis module is operated in a cyclic rolling manner, and the analytical travel time-space path formed by the preliminary analysis module is obtained.

Preferably, if the travel note text of the second travel note sample contains travel photos, interpret the time information and latitude and longitude information in the travel photos, arrange the travel photos based on time order, and generate the image spatiotemporal path of the above travel photos.

Further, part of the travel text is extracted from the second travel sample, and the above-mentioned travel text is tagged with time keywords and location keywords to obtain the marked travel space-time path, and the marked travel space-time path of the travel text is analyzed by the preliminary analysis module. The travel time-space path is compared, based on the differences in the comparison results, the consistency rate of the two is calculated, and the preliminary analysis module is improved.

Judging whether the consistency rate reaches the preset threshold, if the consistency rate does not reach the preset threshold, continue to improve the preliminary analysis module; when the consistency rate reaches the preset threshold, fix the analysis module to obtain the final analysis module.

S40: Using the final parsing module to apply the analytical travel spatio-temporal path in the travel note text of the travel note sample in the preset time window and preset destination range, based on the above-mentioned parsing of the travel spatio-temporal path, construct a travel note spatio-temporal database based on travel notes.

Preferably, collect comment data of tourist attractions, calculate the proportion of comment data of a single tourist attraction in all tourist attractions in the local city, and obtain the number of tourists visiting tourist attractions based on the proportion of comment data;

Count the number of tourist visitors in the check-in spatio-temporal database and the travel notes spatio-temporal database, and calculate the deviation ratio from the number of tourist visitors based on the review data.

Specifically, look up the "Statistical Yearbook" or tourism statistical data released by the official website of tourism data, and obtain the number of tourist receptions in the target city within the time window; collect tourists' comment data on tourist attractions on authoritative travel websites, and obtain The proportion of comment data of all tourist attractions in the local city; multiply the proportion of comment data of each tourist attraction by the number of tourist receptions in the city to obtain the number of tourists visiting a single tourist attraction.

Summarize the data in the check-in spatio-temporal database and the travel notes spatio-temporal database in the same time window and the same area to obtain the number of tourist visitors based on the check-in spatio-temporal database and the travel notes spatio-temporal database, and compare it with the number of tourist visitors obtained from the above-mentioned review data. Compare and calculate the deviation ratio.

若偏差比例y₁小于或等于10％，则证明签到时空数据库的数据源达到饱和，若偏差比例y₁大于10％，则证明签到时空数据库的数据源未饱和，需要增加签到时空数据库采集的数量，直到偏差比例y₂小于或等于10％。Set the number of tourist visitors obtained from the review data as x ₁ , and the number of tourist visitors obtained from the check-in spatio-temporal database as x ₂ , the deviation ratio between the two is

If the deviation ratio y ₁ is less than or equal to 10%, it proves that the data source of the check-in spatio-temporal database is saturated; if the deviation ratio y ₁ is greater than 10%, it proves that the data source of the check-in spatio-temporal database is not saturated, and the number of check-in spatio-temporal databases needs to be increased , until the deviation ratio y2 is less _than or equal to 10%.

若偏差比例y₂小于或等于10％，则证明游记时空数据库的数据源达到饱和，若偏差比例y₂大于10％，则证明游记时空数据库的数据源未达到饱和，需增加游记时空数据库的数据源数量，直到偏差比例y₂小于或等于10％。生成可视化分析之前，对数据库进行饱和性和准确性评估，确定个体数据的可靠性。Assuming that the number of tourist visitors obtained through the spatio-temporal database of travel notes is x ₃ , then the deviation ratio between the number of tourist visitors obtained through the review data is x ₁ is

If the deviation ratio y ₂ is less than or equal to 10%, it proves that the data source of the travel notes spatio-temporal database is saturated; if the deviation ratio y ₂ is greater than 10%, it proves that the data source of the travel notes spatio-temporal database is not saturated, and the data of the travel notes spatio-temporal database needs to be increased The number of sources until the deviation ratio y2 is less _than or equal to 10%. Perform a saturation and accuracy assessment of the database to determine the reliability of individual data before generating visual analytics.

At the same time, the consistency of the two can be judged by comparing the time deviation and location deviation in the analytic travel time-space path formed by the marked travel time-space path and the preliminary analysis module.

After data saturation and accuracy evaluation, a structured database of tourist spatio-temporal behavior in a specified time window and specified area can be obtained.

S50: Construct a tourist flow behavior database based on the check-in spatio-temporal database and the travel notes spatio-temporal database, and obtain a visualized tourist spatio-temporal behavior path map based on the tourist flow behavior database.

Specifically, using the check-in spatio-temporal database and the travel notes spatio-temporal database as the basis, through the "track intervals into lines" tool under the "tracking analyst tools" in the ArcGIS software, the points are connected into lines to form a route map of single or multiple tourists;

The time axis and map base map are displayed together, and finally a visualized tourist spatiotemporal behavior path map is formed for subsequent professional analysis.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented through computer programs to instruct related hardware, and the computer programs can be stored in a non-volatile computer-readable memory In the medium, when the computer program is executed, it may include the processes of the embodiments of the above-mentioned methods. Wherein, any references to memory, storage, database or other media used in the various embodiments provided in the present application may include non-volatile and/or volatile memory. Nonvolatile memory can include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Chain Synchlink, DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

Those skilled in the art can clearly understand that for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used for illustration. In practical applications, the above-mentioned functions can be assigned to different functional units, Module completion means that the internal structure of the system described in this application is divided into different functional units or modules to complete all or part of the functions described above.

Claims (10)

1. A method for extracting guest behavior data, the method comprising:

acquiring tourist attraction sign-in data, and performing structured processing on the tourist attraction sign-in data to obtain a sign-in time-space database based on the tourist attraction sign-in data;

acquiring a first travel note sample from a travel website, marking time information and location information of each travel note text in the first travel note sample to obtain a marked travel space-time path, and forming a preliminary analysis module based on a marking method for marking the travel space-time path;

acquiring a second travel note sample, operating the primary analysis module to obtain analysis travel space-time paths of all travel note texts of the second travel note sample, and perfecting the primary analysis module based on the analysis travel space-time paths to obtain a final analysis module;

applying the final analysis module to travel record samples of a preset time window and a preset destination range to obtain a travel record time-space database based on travel records;

and constructing a tourist movement behavior database based on the check-in time space database and the travel time space database, and obtaining a visualized tourist time space behavior path diagram based on the tourist movement behavior database.

2. The method for extracting tourist behavior data according to claim 1, wherein the step of constructing the tourist movement behavior database based on the check-in time space database and the tourist timing time space database, and obtaining the visualized tourist time space behavior path map based on the tourist movement behavior database further comprises:

collecting comment data of tourist attractions, calculating the comment data proportion of all tourist attractions of a local city in a single tourist attraction, and acquiring the reference number of tourist visions of the tourist attractions based on the comment data proportion;

and obtaining a first number of tourism visiting persons based on the check-in time-space database, obtaining a second number of tourism visiting persons based on the travel time-space database, and respectively calculating the deviation proportion of the first number of tourism visiting persons, the second number of tourism visiting persons and the reference number of tourism visiting persons.

3. The method as claimed in claim 2, wherein after calculating the deviation ratio between the first and second numbers of visitors and the reference number of visitors respectively, the method further comprises:

comparing the deviation proportion with a preset deviation proportion, and if the deviation proportion exceeds the preset deviation proportion, further perfecting the sign-on time-space database and the travel time-space database;

and if the deviation ratio is within a preset deviation ratio, obtaining a tourist flow behavior database on the basis of the check-in time space database and the tour record time space database.

4. The method as claimed in claim 1, wherein the step of obtaining the tourist attraction check-in data and performing the structuring process on the tourist attraction check-in data to obtain the check-in space-time database based on the tourist attraction check-in data comprises:

acquiring a scenic spot sign-in ID of each scenic spot name according to the scenic spot names in the target area to obtain a list of the scenic spot names, the scenic spot sign-in IDs and scenic spot numbers;

acquiring sign-in data of user IDs corresponding to all the scenic spot sign-in IDs of the scenic spots in a time window to obtain an initial user database of the scenic spots, wherein the initial user database comprises the user IDs, sign-in time, sign-in places and sign-in contents;

and acquiring personal information of all user IDs in the initial user database, supplementing the initial user database as an attached table, and acquiring a check-in time space database of the check-in data of the tourist attraction.

5. The guest behavior data extraction method of claim 4, wherein the step of obtaining the guest check-in ID for each guest name further comprises:

and establishing a master check-in ID and a plurality of slave check-in IDs of the scenic spots, summarizing the plurality of slave check-in IDs to the master check-in ID, and using the master check-in ID as the scenic spot check-in ID of each scenic spot.

6. The method according to claim 4 or 5, wherein the step of acquiring check-in data of user IDs corresponding to check-in IDs of all the scenic spots within the time window further comprises:

if the same user ID corresponds to a plurality of different scenic spot sign-in IDs, associating the user ID with the plurality of different scenic spot sign-in IDs;

and if the same user ID corresponds to a plurality of same scenic spot sign-in IDs, performing duplicate removal processing on the same scenic spot sign-in IDs.

7. The method as claimed in claim 1, wherein the step of obtaining a first travel record sample from a travel website, marking time information and location information of each travel record text in the first travel record sample to obtain a marked travel spatiotemporal path, and forming a preliminary analysis module based on the marking method of the marked travel spatiotemporal path comprises:

marking a time keyword of each travel note text in the first travel note sample based on a precise date, a precise time, a fuzzy time and a relative time, placing the time keyword into a time word bank, and dividing the travel note text into text segments by taking the precise date and the precise time as dividing points;

identifying a place keyword for the text segment based on the precise place, the fuzzy place, and the associated place, and placing the place keyword into a place thesaurus;

and constructing a preliminary analysis module based on the time keyword and the place keyword extraction method.

8. The method for extracting tourist behavior data according to claim 1, wherein the step of obtaining a second travel note sample, sequentially operating the preliminary analysis module to obtain analysis travel space-time paths of all travel note texts of the second travel note sample, and completing the preliminary analysis module based on the analysis travel space-time paths to obtain a final analysis module comprises:

and if the second travel record sample contains the travel photos, extracting time information and longitude and latitude information in the travel photos, arranging the travel photos based on a time sequence, and generating an image space-time path of the travel photos.

9. The method as claimed in claim 1, wherein the step of refining the preliminary analysis module based on the analysis of the travel spatiotemporal path to obtain the final analysis module comprises:

and extracting part of travel note texts from the second travel note samples, marking the travel note texts to obtain marked travel space-time paths, comparing the analyzed travel space-time paths obtained by the preliminary analysis module with the marked travel space-time paths, and perfecting the preliminary analysis module based on a comparison result to obtain a final analysis module.

10. The method of claim 1, wherein the step of obtaining a visualized tourist spatiotemporal behavior path map based on the tourist flow behavior database comprises:

and forming a path map of single or multiple tourists through an ArcGIS software tool based on the tourist flow behavior database, and adding a time axis and a map base map into the path map to form a visualized tourist space-time behavior path map.

Images