Tourist attraction building feature recommendation method and device based on big data analysis

Abstract

The invention provides a scenic spot building feature recommendation method and device based on big data analysis, and belongs to the field of scenic spot recommendation. The problem that specific buildings of scenic spots are not recommended in existing scenic spot recommendation is solved; the method specifically comprises the following steps: S1, acquiring user information and scenic spot information; s2, processing the scenic spot information and the user information; obtaining a correlation value deviation, a scenic spot geographic position weight value and a scenic spot recommendation weight value; s3, combining the correlation value deviation, the geographic position weight value of the scenic spot and the recommendation weight value to obtain recommendation deviation of the scenic spot; obtaining recommended scenic spots according to the recommendation deviation; s4, analyzing the buildings in the recommended scenic spot to obtain a building recommendation value, obtaining a recommendation judgment value according to the building recommendation value and the number of the buildings, and recommending the buildings according to the judgment value; according to the invention, by analyzing the user preference, the scenic spots are recommended for the user, the buildings in the scenic spots are specifically analyzed, and the specific buildings are recommended for the user.

Description

Technical Field

[0001] The present invention discloses a method and a device for recommending architectural features of tourist attractions based on big data analysis, and relates to the field of tourist attraction recommendation. Background Art

[0002] The existing methods and devices for recommending tourist attraction buildings have the following deficiencies:

[0003] Scenic spot recommendations are mostly focused on the scenic area level: most scenic spots are recommended to users, and there are no clear recommendations for various types of buildings in the scenic area. When users want to visit related buildings, they cannot choose scenic spots through existing recommendations.

[0004] Few recommendation factors: Attraction recommendations are mainly based on the popularity of the attractions, and the recommended content is the same for different users. The personalized needs of users are ignored, and specific recommendations cannot be made based on user needs.

[0005] Privacy and security risks: The system involves the collection and transmission of user personal information and scenic spot information, which poses certain privacy and security risks and requires strengthening of data protection and privacy protection measures. Summary of the invention

[0006] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a method and device for recommending tourist attraction buildings based on big data analysis, aiming to solve the problem of unreasonable tourist attraction recommendations.

[0007] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a method for recommending architectural features of tourist attractions based on big data analysis, the recommendation method comprising:

[0008] Step S1: Obtain the user's geographical location, hobbies, and browsing history to obtain user information, and obtain the name of the scenic spot, scenic spot characteristics, scenic spot architectural style, scenic spot historical and cultural background, and scenic spot traffic information to obtain scenic spot information;

[0009] Step S2: Processing scenic spot information and user information; counting scenic spot feature keywords according to scenic spot features in the scenic spot information, searching for scenic spot feature keywords through interests and browsing records in the user information, obtaining user preferences, combining user preferences with scenic spot information, obtaining correlation values ​​between scenic spots and user preferred scenic spots, and calculating correlation value deviations of scenic spots;

[0010] The geographical location of the scenic spot is obtained according to the name of the scenic spot, and the relative distance value from the user to each scenic spot is obtained from the geographical location of the user and the geographical location of the scenic spot, and the weight value of the geographical location of the scenic spot is obtained from the relative distance value;

[0011] The recommended weight value of the scenic spot is obtained by analyzing and calculating the architectural style, historical and cultural background, and visitor flow information of the scenic spot;

[0012] Step S3: combining the correlation value deviation, the geographical location weight value and the recommendation weight value to obtain the recommended deviation of the scenic spot; obtaining the recommended scenic spot based on the recommended deviation;

[0013] Step S4: According to the recommended scenic spots, the architectural information of the scenic spots is obtained, the architectural information of the scenic spots is analyzed to obtain the building recommendation value, the building recommendation judgment value is obtained from the building recommendation value and the building information, and the characteristic buildings are recommended according to the building recommendation value and the building recommendation judgment value.

[0014] Furthermore, the specific steps of step S2 are as follows:

[0015] Step S21: Obtain scenic spot feature keywords through scenic spot features in the scenic spot information, and count all scenic spot feature keywords;

[0016] Step S22: taking the scenic spot information of all scenic spots as a set, denoted as D, and counting the number of scenic spots |D| in the set D; obtaining the number of scenic spot architectural styles according to the scenic spot information, and counting the number of scenic spots of each architectural style to obtain the number of scenic spots of each architectural style;

[0017] Step S23: Obtaining a recommended value for a scenic spot based on the historical and cultural background of the scenic spot and the age group of the user;

[0018] Step S24: retrieve the scenic spot feature keywords according to the interests and browsing records in the user information to obtain the user preferences, retrieve the user preferences, obtain the frequency of occurrence of the user preferences in the user information, and obtain the top sl user preferences with the highest frequency; put these preferences in the same set as the user's preference set; the number of preferences in the preference set is sl;

[0019] Step S25: obtaining a correlation value between the scenic spot and the user preference according to the user preference and the scenic spot information, and calculating the correlation value deviation based on the correlation value;

[0020] Step S26: Obtain the geographical location of the scenic spot according to the name of the scenic spot, and obtain the relative distance |x|j between the geographical location of the user and the geographical location of the scenic spot in combination with the geographical location of the user in the user information; calculate the ratio of the distance between each scenic spot and the user according to the relative distance, and obtain the weight value of the geographical location of the scenic spot, which is recorded as f(|x|j);

[0021] The specific calculation process of the weight value of the scenic spot's geographical location is as follows:

[0022]

[0023] Where: |D| is the number of scenic spots, 0 <j≤|D|;

[0024] Step S27: Obtain a first weight based on the tourist flow information of the scenic spot in the past three years; obtain a second weight by combining the first weight and the proportion of the number of scenic spots of each architectural style in the number of all scenic spots; and obtain a recommended weight value by combining the second weight and the scenic spot recommendation value.

[0025] Furthermore, the specific steps of step S23 are as follows:

[0026] Step S231: Obtain the number of users of different age groups who like b dynasties cd1, cd2, cd3, ... cdb through a questionnaire survey;

[0027] Step S232: Calculate the proportion of the number of people who like each dynasty in the total number of people to obtain the proportion of the number of people who like each dynasty, denoted as zb;

[0028] The specific calculation of the population ratio is as follows:

[0029]

[0030] Among them: cdx represents the number of users in the age group who like the xth dynasty, and the value range of x is: 1≤x≤b;

[0031] Step S233: sort the population ratio values ​​in descending order; obtain the rank of each scenic spot in the ranking, and obtain the ranking value cw; obtain the historical and cultural background of the scenic spot according to the scenic spot information, obtain the historical figures related to each scenic spot in the historical and cultural background of the scenic spot, and obtain the number of historical figures gs associated with the scenic spot, and obtain the recommended value of the scenic spot in combination with the ranking value cw, which is recorded as tjz;

[0032] The calculation steps for the recommended value of scenic spots are as follows:

[0033]

[0034] Furthermore, the specific steps of step S25 are as follows:

[0035] Step S251: according to the scenic spot information, obtain the number T of all words in each scenic spot information; traverse the preference set, obtain the number of occurrences t of the user's preference in each scenic spot information, and combine the number T of all words in the scenic spot information to obtain the word frequency TF of the user's preference;

[0036] The word frequency calculation process of user preference is as follows:

[0037]

[0038] Obtain the number of attractions that contain user preferences in all attraction information and calculate the inverse document frequency;

[0039]

[0040] Where |D| is the number of scenic spots, and g(i) represents the number of scenic spots preferred by the i-th user in the preference set;

[0041] Obtain the correlation value p between user preference and scenic spot based on the user's preferred word frequency TF and inverse document frequency IDF;

[0042] The calculation process of the correlation value p between user preferences and attractions is as follows:

[0043] p = TF × IDF;

[0044] Count all the related values ​​and get the matrix A;

[0045]

[0046] Among them, p11 refers to the correlation value between the first user preference and the first scenic spot in the preference set, and so on, pij refers to the correlation value between the i-th user preference and the j-th scenic spot in the preference set;

[0047] Step S252: Calculate the correlation value of each preference in the preference set to obtain the correlation value deviation of the scenic spot, recorded as Ej;

[0048] The specific calculation steps of the correlation value deviation Ej are as follows:

[0049]

[0050] Where pij is the corresponding value in the matrix A, sl is the number of preferences in the preference set, 1≤i≤sl.

[0051] Furthermore, the specific steps of step S27 are as follows:

[0052] Step S271: according to the tourist flow information of the scenic spot, obtain the tourist flow num1, num2, num3 of the scenic spot in the past three years; obtain the first weight according to the tourist flow of the scenic spot in the past three years;

[0053] Step S272: Obtain the number of scenic spots of each architectural style and the number of all scenic spots, calculate the proportion of the scenic spot architectural style, obtain the proportion value of the scenic spot architectural style, and obtain the second weight by combining the proportion value with the first weight;

[0054] Step S273: combining the scenic spot recommendation value and the second weight to obtain a recommendation weight value, denoted as kk;

[0055] The specific calculation process of the recommended weight value is as follows:

[0056] kk=k2×tjz;

[0057] Among them, k2 is the second weight, and tjz is the recommended value of the scenic spot.

[0058] Furthermore, the specific steps of step S271 are as follows:

[0059] Step S2711: Calculate the average number of visitors to the scenic spot based on the number of visitors to the scenic spot in the past three years, and record it as year(dj);

[0060] The average visitor flow of an attraction is calculated as follows:

[0061]

[0062] Among them: dj represents a specific scenic spot; num1, num2, and num3 represent the current year's passenger flow calculated one, two, and three years ago respectively from the current year;

[0063] Step S2712: Obtain the average value of the scenic spots by combining the average value of the flow of people at the scenic spots with the number of scenic spots, recorded as all;

[0064] The average value of the scenic spot is calculated as follows:

[0065]

[0066] Among them: year represents the average flow of tourists to the scenic spot, |D| is the number of scenic spots, 0 <j≤|D|;

[0067] Step S2713: Obtain the first recommendation weight according to the average flow rate of the scenic spot, the average flow rate of the scenic spot, and the flow rates num1 and num3, denoted as k1:

[0068] The specific calculation process of the first recommendation weight is as follows:

[0069]

[0070] Furthermore, the specific steps of step S272 are as follows:

[0071] Step S2721: by calculating the number of scenic spots with this architectural style and the number of all scenic spots, a proportion of the scenic spot architectural style is obtained, which is recorded as bl;

[0072] The calculation process of the proportion of scenic spot architectural style is as follows:

[0073]

[0074] Where: Gu is the number of scenic spots with the u-th architectural style, |D| is the number of all scenic spots;

[0075] Step S2722: Calculate the proportion of the architectural style of the scenic spot and the first weight to obtain a second weight, recorded as k2;

[0076] The specific calculation process of the second weight is as follows:

[0077] k2=k1×bl;

[0078] Among them: k1 is the first weight.

[0079] Furthermore, the specific steps of step S3 are as follows:

[0080] Step S31: Obtain the recommended deviation of the scenic spot according to the correlation value deviation, the geographical location weight value, and the recommended weight value, which is recorded as y;

[0081] The recommended deviation is calculated as follows:

[0082]

[0083] Where: f(|x|j) is the geographical location weight value, Ej is the correlation value deviation, kk is the recommended weight value, 0 <j≤|D|;

[0084] Step S32: sort the recommended deviations in descending order, and obtain the scenic spot with the smallest recommended deviation as the recommended scenic spot.

[0085] Furthermore, the specific steps of step S4 are as follows:

[0086] Step S41: Obtain the number of buildings in the recommended scenic spot, the number of deeds of people related to the buildings, the number of visitors and the duration of the visit to obtain building information, and obtain the proportion of deeds of people related to each building from the number of buildings in the recommended scenic spot and the number of deeds of people related to the buildings, recorded as hc;

[0087] The proportion of building-related deeds is calculated as follows:

[0088]

[0089] Among them: jz is the number of buildings in the scenic spot, rwc is the number of deeds of people related to the buildings, <c≤jz;

[0090] Step S42: according to the building information of the recommended scenic spot, the number of visitors and the visiting time of each building in the past three months are obtained; the total visiting time of tourists in each building is obtained from the number of visitors and the visiting time of each building, which is recorded as zcc;

[0091] The total visit time of each building is calculated as follows:

[0092]

[0093] Among them: rsc is the number of visitors to each building in the past three months, sccr is the length of time that visitors spend visiting each building in the past three months;

[0094] Step S43: according to the building information of the recommended scenic spot, the number of visitors and the visiting time of each building in the previous month are obtained; according to the total visiting time of tourists, the number of visitors and the visiting time of each building in the previous month, the visiting time change value of the building is obtained, which is recorded as bhc;

[0095] The change in the building's visit duration is calculated as follows:

[0096]

[0097] Among them: dsc is the number of visitors to each building in the adjacent month, dccr is the visit time of tourists in each building in the adjacent month;

[0098] Step S44: Obtain the building recommendation value based on the total visit time zcc of tourists, the proportion of building-related figures and deeds hc, and the visit time change value bhc, which is recorded as tjc;

[0099] The specific calculation of the building recommended value is as follows:

[0100] tjc=zcc×hc×bhc;

[0101] Step S45: Obtain a building recommendation judgment value, denoted as pj, from the building recommendation value tjc and the number of buildings in the scenic spot jz;

[0102] The specific calculation process of the building recommendation judgment value is as follows:

[0103]

[0104] Step S46: Compare the building recommendation value with the building recommendation judgment value, and take the building whose building recommendation value is greater than the building recommendation judgment value as the characteristic building of the scenic spot, and recommend the characteristic building to the user.

[0105] A device for recommending architectural features of tourist attractions based on big data analysis, the recommendation device comprising:

[0106] Collection module: used to obtain the user's geographic location, hobbies, and browsing history to obtain user information, and obtain the name of the scenic spot, the characteristics of the scenic spot, the architectural style of the scenic spot, the historical and cultural background of the scenic spot, and the flow of people in the scenic spot to obtain the scenic spot information;

[0107] Data processing module: used to process scenic spot information and user information; obtain user preferences through user information, combine user preferences with scenic spot information, obtain the correlation value between scenic spots and user preferred scenic spots, and calculate the correlation value deviation of scenic spots;

[0108] Obtain the geographical location of the scenic spot according to the scenic spot name. Based on the geographical location of the user and the scenic spot, obtain the relative distance value from the user to each scenic spot, and obtain the scenic spot geographical location weight value from the relative distance value; analyze and calculate the architectural style of the scenic spot, the historical and cultural background of the scenic spot, and the passenger flow information of the scenic spot to obtain the recommended weight value of the scenic spot;

[0109] Calculation module: used to calculate the relevant value deviation, the scenic spot geographical location weight value and the recommended weight value, obtain the recommended deviation of the scenic spot, and obtain the recommended scenic spot from the recommended deviation;

[0110] Recommendation module: According to the recommended scenic spot, obtain the number of buildings in the scenic spot, the number of deeds of relevant figures of the buildings, the number of visitors and the visiting duration. Analyze the number of deeds of relevant figures of the buildings, the number of visitors and the visiting duration to obtain the building recommendation value. According to the building recommendation value and the number of buildings in the scenic spot, obtain the building recommendation judgment value, and conduct the recommendation of characteristic buildings according to the building recommendation value and the building recommendation judgment value.

[0111] Compared with the prior art, the beneficial effects of the present invention are:

[0112] Refine the recommended content: The device further analyzes the recommended scenic spot and recommends the buildings in the scenic spot. Sort the characteristic buildings in the scenic spot through the deeds of relevant figures of the buildings and the visiting duration of tourists, and recommend the characteristic buildings to the user.

[0113] Meet the personalized needs of users: The device combines the user's interests and hobbies, the user's browsing situation through the recommendation module, makes corresponding predictions on the scenic spots and buildings that the user wants to go to, and conducts personalized recommendations for the user.

[0114] Combine multiple factors: The device analyzes the user's preferences through the information processing module, combines the user's geographical location, the historical and cultural background of the scenic spot, and the passenger flow information, and conducts detailed processing and analysis of the recommended scenic spot through multiple factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0115] By reading the detailed description of the non-restrictive embodiments with reference to the following drawings, other features, purposes and advantages of the present invention will become more obvious:

[0116] Figure 1 It is a schematic diagram of the system flow of the present invention;

[0117] Figure 2 It is a schematic diagram of the method flow of the present invention;

[0118] Figure 3 It is a schematic diagram of the distance between the scenic spot and the user of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0119] The technical solution of the present invention will be described clearly and completely in conjunction with the embodiments below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0120] See also Figure 1 The device for recommending architectural features of tourist attractions based on big data analysis includes: an acquisition module, a data processing module, a calculation module, a recommendation module, and a database; the acquisition module, the data processing module, the calculation module, and the recommendation module are respectively connected to the database.

[0121] The database stores the user's geographic location, hobbies, browsing history, and the name, characteristics, architectural style, historical and cultural background, and visitor flow information of the scenic spot.

[0122] The workflow of the acquisition module is as follows:

[0123] The collection module accesses the database to read the user's geographical location, hobbies, and browsing history in the database as user information;

[0124] The acquisition module accesses the database to read the name, characteristics, architectural style, historical and cultural background, and visitor flow information of the scenic spot in the database as the scenic spot information. The information of all scenic spots is taken as a set and recorded as D.

[0125] The data processing module processes the scenic spot information and user information. The workflow is as follows:

[0126] The data processing module processes the scenic spot information:

[0127] Obtain keywords of scenic spot characteristics through scenic spot characteristics, such as mountains, ancient buildings, and science and technology museums; and count all keywords of scenic spot characteristics;

[0128] Keyword examples:

[0129] Natural scenery: mountains, lakes, oceans, grasslands, organisms...

[0130] Historical and cultural: ancient architecture, relics, memorial halls, folklore, legends...

[0131] Architectural landmarks: bridges, tunnels, high-rise buildings...

[0132] Science and technology innovation: science and technology exhibitions, science and technology museums, creative parks...

[0133] The number of scenic spots |D| in the set D is counted; the number of scenic spot architectural styles is obtained according to the scenic spot information, and the number of scenic spots of each architectural style is counted to obtain the number of scenic spots of each architectural style.

[0134] It should be noted that architectural style refers to the unique design features displayed by the buildings in the scenic spot, such as Buddhist temple architecture, Chinese garden architecture, and Taoist temple architecture.

[0135] Sort scenic spots from different dynasties by user age group and historical and cultural background;

[0136] The specific process of sorting the attractions is as follows:

[0137] The age groups are divided into the first age group to the nth age group, where n≥3. The dynasty preferences of all age groups are obtained through a questionnaire survey (the number of respondents is not less than 10,000, and includes all age groups and different rich and poor areas). The number of people who like dynasties in the first age group to the nth age group is counted respectively, and the number of people in the first age group is a1, the number of people in the second age group is a2... The number of people in the nth age group is an. The user age is obtained, and the user age group is determined according to the user age. The number of people who like each dynasty in the user age group is counted. The number of dynasties is set to b, and the number of people in the first dynasty is cd1, the number of people in the second dynasty is cd2... The number of people in the bth dynasty is cdb.

[0138] The proportion of people in each dynasty is calculated as follows:

[0139]

[0140] Among them: cdx represents the number of users in the age group who like the xth dynasty, and the value range of x is: 1≤x≤b;

[0141] Sort in descending order based on the percentage of people to obtain the ranking of the number of people in different dynasties that users in this age group like.

[0142] Obtain the ranking of each scenic spot in the ranking, and obtain the ranking value cw; obtain the historical and cultural background of the scenic spot according to the scenic spot information, obtain the historical figures related to each scenic spot in the historical and cultural background of the scenic spot, and obtain the number of historical figures associated with the scenic spot gs, and combine the ranking value cw and the number of historical figures associated with the scenic spot gs to obtain the scenic spot recommendation value, recorded as tjz;

[0143] The calculation steps for the recommended value of scenic spots are as follows:

[0144]

[0145] It should be noted that: the historical figures associated with the scenic spots refer to those who are well-known and have a relationship with the scenic spots; for example, for the Zhuge Cottage in Nanyang, Li Bai and Liu Yuxi wrote poems about it, and Zhuge Liang lived there, so the number of figures associated with the scenic spot is 3.

[0146] The data processing module processes the user information, matches the feature keywords of the scenic spots by combining the user's interests and browsing records, and obtains the user's scenic spot keywords. The specific process is as follows:

[0147] Obtain the user's interests and browsing records in the past six months;

[0148] Use the text cleaning tool to clean the data of the user's interests and browsing records, removing expressions and numbers;

[0149] Use the cleaned data to perform word frequency statistics on the feature keywords. Input the cleaned data into the word frequency processing software, and set the keywords as the search terms; obtain the word frequency information about the keywords;

[0150] Obtain the top sl keywords with the most word frequencies;

[0151] Put these keywords in the same set as the user's preference set.

[0152] It should be noted that: the website of the text cleaning tool: https: / / www.txttool.com / ; the word frequency processing software is Micro Word Cloud; preference refers to the tendency shown by an individual or a group in making choices or decisions. In the present invention, it mainly refers to the influencing factors that make the user choose a scenic spot when choosing a scenic spot. The preference set is a set of these influencing factors that affect the user's choice of scenic spots.

[0153] The data processing module matches the keywords in the user's preference set with the scenic spot information. The specific matching steps are as follows:

[0154] According to the scenic spot information, obtain the number of all words T in each scenic spot information; traverse the preference set, obtain the number of occurrences t of the user's preference in each scenic spot information, and combine the number of all words T in the scenic spot information to obtain the word frequency TF of the user's preference;

[0155] The calculation process of the word frequency of the user's preference is as follows:

[0156]

[0157] It should be noted that i represents the i-th preference in the preference set, 0 < i <= sl; D(j) represents the j-th scenic spot information in the scenic spot information set, 0 < j <= |D|; t represents the number of occurrences of the i-th preference in the preference set in the scenic spot information D(j), and T represents the number of all words in the scenic spot information D(j).

[0158] Obtain the number of attractions that contain user preferences in all attraction information, and calculate the inverse document frequency IDF;

[0159]

[0160] Where |D| is the total number of scenic spot information, and g(i) represents the number of scenic spot information containing the i-th preference in the preference set.

[0161] It should be noted that inverse document frequency is a weighting technique commonly used in information retrieval and text mining. It reduces the weight of common words and increases the weight of rare words through inverse document frequency, thereby improving the accuracy of text analysis.

[0162] Obtain the correlation value p between user preference and scenic spot based on the occurrence frequency TF of user preference in each scenic spot information and the inverse document frequency IDF of preference;

[0163] p = TF × IDF;

[0164] Save all the relevant values ​​of user preferences and attractions to obtain matrix A;

[0165]

[0166] Extract the relevant value of each scenic spot in the matrix, perform operations on each column in the matrix, and obtain the relevant value deviation Ej of the user for each scenic spot;

[0167]

[0168] Ej represents the deviation of the correlation value for the jth scenic spot; pij is the correlation value of the jth scenic spot with respect to the i-th keyword in the preference set;

[0169] For example, for the first attraction:

[0170]

[0171] The second attraction:

[0172]

[0173] Similarly, the correlation value deviations of all |D| scenic spots are obtained;

[0174] See also Figure 3 , obtain the geographical location of the scenic spot according to the name of the scenic spot, and combine the geographical location of the user in the user information to obtain the relative distance |x|j between the geographical location of the user and the geographical location of the scenic spot; calculate the ratio of the distance between each scenic spot and the user according to the relative distance, and obtain the weight value of the geographical location of the scenic spot, recorded as f(|x|j);

[0175]

[0176] It should be noted that the relative distance between the scenic spot and the user refers to the straight-line distance obtained through the map.

[0177] The data processing module calculates the scenic spot information and obtains the recommended weight value;

[0178] The specific calculation process is as follows:

[0179] According to the tourist flow information of the scenic spot, the tourist flow of the scenic spot in the past three years is obtained and recorded as num1, num2, and num3 respectively.

[0180] The average number of visitors to the scenic spot in the past three years, num1, num2, and num3, is used to obtain the average number of visitors to the scenic spot year;

[0181]

[0182] It should be noted that: dj represents a specific scenic spot, dj∈D; num1, num2, num3 represent the annual passenger flow data calculated one year, two years, and three years forward from the current year respectively; if the passenger flow information is less than three years, the average passenger flow information of all years is calculated.

[0183] The average value of all attractions is obtained by the average year of the flow of tourists to the attractions and the number of attractions |D|;

[0184]

[0185] The first weight k1 is obtained based on the average year of the scenic spot’s flow, the average all of the scenic spot’s flow, and the flow of people num1 and num3. The calculation process of k1 is as follows:

[0186]

[0187] It should be noted that for attractions with less than three years of traffic information, if there are only two years, only num1 / num2 is used to obtain the traffic change; if there is only one year, it is represented by 1. For attractions with only one year of traffic information, we only judge whether to recommend it based on the traffic volume of that year.

[0188] The proportion of scenic spot architectural style bl is obtained by the number of scenic spots of this architectural style Gu and the number of all scenic spots |D|;

[0189]

[0190] Where: Gu is the number of scenic spots with the u-th architectural style, 1≤u≤v, |D| is the number of all scenic spots;

[0191] The second weight k2 is obtained by the proportion value bl of the architectural style of the scenic spot and the first weight k1;

[0192] k2=k1×bl;

[0193] According to the scenic spot recommendation value tjz and the second weight k2, a recommendation weight value kk is obtained;

[0194] kk=k2×tjz;

[0195] The calculation module obtains the recommended deviation y of the scenic spot according to the relevant value deviation Ej, the geographical location weight value f(|x|j), and the recommended weight value kk;

[0196]

[0197] Where f(|x|j) is the ratio of the relative distance between the user and the selected scenic spot, Ej is the deviation between the scenic spot and the user's preference, and kk is the recommendation weight value obtained based on the scenic spot flow information, scenic spot architectural style, and scenic spot historical and cultural background;

[0198] The recommendation module provides recommended attractions to users based on the recommended deviation y of the attractions, arranges the y values ​​in descending order, and recommends the attractions with the smallest y values ​​to users as recommended attractions;

[0199] According to the recommended attractions, obtain the number of buildings in the attractions, the number of people's deeds related to the buildings, the number of visitors and the duration of the visit;

[0200] According to the number of buildings jz in the scenic spot and the number of deeds of people related to the buildings rwc, the proportion value hc of the deeds of people related to each building is obtained;

[0201]

[0202] rwc represents the number of deeds related to the cth building in the scenic spot, hc is the ratio of the number of deeds related to the cth building to the number of deeds related to all buildings, 0 <c<=jz;

[0203] It should be noted that the deeds of people related to buildings refer to the things that happened to famous people in history at that place. For example, Du Fu wrote "Climbing Yueyang Tower" in Yueyang Tower, and Bai Juyi wrote "Remembering Yueyang Tower" when he passed by Yueyang Tower. These are two deeds of people.

[0204] Obtain the number of visitors and visiting duration of each building in the recommended scenic spots in the past three months, and the number of visitors and visiting duration of each building in the recommended scenic spots in the next month;

[0205] It should be noted that the past three months refer to the three months before this month, and the adjacent month refers to the previous month. For example, if this month is December, the previous three months are September, October, and November, and the adjacent month is November.

[0206] According to the number of visitors and the duration of visits in the past three months, the number of visitors to each building (rsc) and the duration of visitors to each building (sccr) are obtained; the total duration of visitors to each building (zcc) is calculated;

[0207]

[0208] sccr refers to the visit time of the rth visitor in the cth building;

[0209] According to the number of visitors and the duration of visits in the adjacent month, the number of visitors to each building (dsc) and the duration of visits to each building (dccr) are obtained;

[0210] According to the total visit time zcc of tourists, the number of visitors to each building in the previous month dsc, and the visit time dccr of tourists in each building, the change value bhc of the visit time this month is obtained;

[0211]

[0212] dc cr Indicates the visiting time of the rth visitor at the cth building.

[0213] The building recommendation value tjc is obtained by the total visit time zcc of tourists, the proportion of building-related deeds hc, and the change value bhc of visit time;

[0214] tjc=zcc×hc×bhc;

[0215] The building recommendation judgment value is obtained by the building recommendation value and the number of buildings in the scenic spot, denoted as pj;

[0216] The specific calculation process of the building recommendation judgment value is as follows:

[0217]

[0218] The building recommendation value is compared with the building recommendation judgment value, and the building with a building recommendation value greater than the building recommendation judgment value is regarded as a characteristic building of the scenic spot, and the characteristic building is recommended to the user.

[0219] It should be noted that when the y value is 0, it means that the user is at the scenic spot and no recommendation is needed.

[0220] See also Figure 2 Based on another concept of the same invention, a method for recommending architectural features of tourist attractions based on big data analysis is proposed, comprising the following steps:

[0221] Step S1: Obtain the user's geographical location, hobbies, and browsing history to obtain user information, and obtain the name of the scenic spot, scenic spot characteristics, scenic spot architectural style, scenic spot historical and cultural background, and scenic spot traffic information to obtain scenic spot information;

[0222] Step S2: Processing scenic spot information and user information; counting scenic spot feature keywords according to scenic spot features in the scenic spot information, searching for scenic spot feature keywords through interests and browsing records in the user information, obtaining user preferences, combining user preferences with scenic spot information, obtaining correlation values ​​between scenic spots and user preferred scenic spots, and calculating correlation value deviations of scenic spots;

[0223] The geographical location of the scenic spot is obtained according to the name of the scenic spot, and the relative distance value from the user to each scenic spot is obtained from the geographical location of the user and the geographical location of the scenic spot, and the weight value of the geographical location of the scenic spot is obtained from the relative distance value;

[0224] The recommended weight value of the scenic spot is obtained by analyzing and calculating the architectural style, historical and cultural background, and visitor flow information of the scenic spot;

[0225] Step S21: Obtain scenic spot feature keywords through scenic spot features in the scenic spot information, and count all scenic spot feature keywords;

[0226] Step S22: taking the information of all scenic spots as a set, denoted as D, counting the number of scenic spots |D| in the set D, obtaining the number of scenic spot architectural styles according to the scenic spot information, and counting the number of scenic spots of each architectural style to obtain the number of scenic spots of each architectural style;

[0227] Step S23: Obtaining a recommended value for a scenic spot based on the historical and cultural background of the scenic spot and the age group of the user;

[0228] The process of obtaining the recommended value of scenic spots is as follows:

[0229] Step S231: Obtain the number of users of different age groups who like b dynasties cd1, cd2, cd3, ... cdb through a questionnaire survey;

[0230] Step S232: by calculating the proportion of the number of people who like each dynasty in the total number of people, the proportion of people who like each dynasty is obtained, which is recorded as zb;

[0231] The specific calculation of the population ratio is as follows:

[0232]

[0233] Among them: cdx represents the number of users in the age group who like the xth dynasty, and the value range of x is: 1≤x≤b;

[0234] Step S233: sort the population ratio values ​​in descending order; obtain the rank of the dynasty in which each scenic spot was established in the sorting, and obtain the sorting rank value cw; obtain the historical and cultural background of the scenic spot according to the scenic spot information, obtain the historical figures related to each scenic spot in the historical and cultural background of the scenic spot, and obtain the number of historical figures associated with the scenic spot, and combine the sorting rank value and the number of historical figures associated with the scenic spot to obtain the scenic spot recommendation value, which is recorded as tjz;

[0235] The calculation steps for the recommended value of scenic spots are as follows:

[0236]

[0237] Where cw is the ranking value, gs is the number of people associated with the scenic spot;

[0238] Historical figures associated with scenic spots refer to people who are related to the scenic spot and are well-known; for example, for Zhuge Lu in Nanyang, Li Bai and Liu Yuxi wrote poems for it and Zhuge Liang lived here, so the number of people associated with the scenic spot is 3;

[0239] Step S24: retrieve the scenic spot feature keywords through the interests and browsing records in the user information to obtain the user preferences, retrieve the user preferences through the word frequency processing software, obtain the frequency of occurrence of the user preferences in the user information, and obtain the first sl user preferences with the highest frequency, sl is a positive integer greater than 0; put these preferences in the same set as the user's preference set; the number of preferences in the preference set is sl;

[0240] Step S25: obtaining a correlation value between the scenic spot and the user preference according to the user preference and the scenic spot information, and calculating the correlation value deviation based on the correlation value;

[0241] The steps for calculating the correlation value deviation are as follows:

[0242] Step S251: according to the scenic spot information, obtain the number T of all words in each scenic spot information; traverse the preference set, obtain the number of occurrences t of the user's preference in each scenic spot information, and combine the number T of all words in the scenic spot information to obtain the word frequency TF of the user's preference;

[0243] The word frequency calculation process of user preference is as follows:

[0244]

[0245] Obtain the number of attractions that contain user preferences in all attraction information and calculate the inverse document frequency;

[0246]

[0247] Where |D| is the number of scenic spots, and g(i) represents the number of scenic spots preferred by the i-th user in the preference set;

[0248] Obtain the correlation value p between user preference and scenic spot based on the user's preferred word frequency TF and inverse document frequency IDF;

[0249] The calculation process of the correlation value p between user preferences and attractions is as follows:

[0250] p = TF × IDF;

[0251] Count all the related values ​​and get the matrix A;

[0252]

[0253] Among them, p11 refers to the correlation value between the first user preference and the first scenic spot in the preference set, and so on, pij refers to the correlation value between the i-th user preference and the j-th scenic spot in the preference set;

[0254] Step S252: Calculate the correlation value of each preference in the preference set to obtain the correlation value deviation of the scenic spot, recorded as Ej;

[0255] The specific calculation steps of the correlation value deviation Ej are as follows:

[0256]

[0257] Where pij is the corresponding value in the matrix A, sl is the number of preferences in the preference set, 1≤i≤sl.

[0258] Step S26: Obtain the relative distance between the user's location and all scenic spots through the user's location in the user information; obtain the ratio of the distance between each scenic spot and the user according to the relative distance, and obtain the weight value of the scenic spot's location, recorded as f(|x|j);

[0259] The specific calculation process of the weight value of the scenic spot's geographical location is as follows:

[0260]

[0261] Among them: |x|j is the relative distance between the user location and the scenic spot location, and |D| is the number of scenic spots.

[0262] Step S27: Obtain a first weight based on the tourist flow of the scenic spot in the past three years; obtain a second weight by combining the first weight and the proportion of the number of scenic spot architectural styles in all scenic spots; obtain a recommended weight value by combining the second weight and the scenic spot recommendation value;

[0263] Step S271: Obtain the tourist flow information num1, num2, and num3 of the scenic spot in the past three years; obtain the first weight according to the tourist flow information of the scenic spot in the past three years;

[0264] Step S2711: Calculate the average passenger flow of the scenic spot based on the passenger flow information of the scenic spot in the past three years, and record it as year(dj);

[0265] The average visitor flow of attractions is calculated as follows:

[0266]

[0267] Among them: dj represents a specific scenic spot; num1, num2, and num3 represent the current year's passenger flow data calculated one, two, and three years ago respectively from the current year;

[0268] Step S2712: Obtain the average value of the scenic spot through the average value of the flow of tourists at the scenic spot and the number of scenic spots, which is recorded as all;

[0269] The average value of the scenic spot is calculated as follows:

[0270]

[0271] Among them: year represents the average flow of tourists to the scenic spot, |D| is the number of scenic spots;

[0272] Step S2713: Obtain the first recommendation weight according to the average flow rate of scenic spots, the average flow rate of scenic spots and the flow rates num1 and num3, denoted as k1:

[0273] The specific calculation process of the first recommendation weight is as follows:

[0274]

[0275] Step S272: Obtain the number of scenic spots of each architectural style and the total number of scenic spots, calculate the proportion of the architectural style of the scenic spot, obtain the proportion value of the architectural style of the scenic spot, and obtain the second weight by combining the proportion value with the first weight;

[0276] Step S2721: by calculating the number of scenic spots with this architectural style and the number of all scenic spots, a proportion of the scenic spot architectural style is obtained, which is recorded as bl;

[0277] The calculation process of the proportion of scenic spot architectural style is as follows:

[0278]

[0279] Where: Gu is the number of scenic spots with the u-th architectural style, |D| is the number of all scenic spots;

[0280] Step S2722: Calculate the proportion of the architectural style of the scenic spot and the first weight to obtain a second weight, recorded as k2;

[0281] The specific calculation process of the second weight is as follows:

[0282] k2=k1×bl;

[0283] Step S273: combining the scenic spot recommendation value and the second weight to obtain a recommendation weight value, denoted as kk;

[0284] The specific calculation process of the recommended weight value is as follows:

[0285] kk=k2×tjz;

[0286] Among them: tjz is the recommended value of the scenic spot.

[0287] Step S3: combining the correlation value deviation, the geographical location weight value and the recommendation weight value to obtain the recommended deviation of the scenic spot; sorting the recommended deviations to obtain the scenic spot with the smallest recommended deviation as the recommended scenic spot;

[0288] The process of obtaining the recommended deviation is as follows:

[0289] Step S31: Obtain the recommended deviation of the scenic spot according to the correlation value deviation, the geographical location weight value, and the recommended weight value, which is recorded as y;

[0290] The recommended deviation is calculated as follows:

[0291]

[0292] Where: f(|x|j) is the geographical location weight value, Ej is the correlation value deviation, and kk is the recommended weight value;

[0293] Step S32: sort the recommended deviations in descending order, and recommend the scenic spot with the smallest recommended deviation to the user as the recommended scenic spot.

[0294] Step S4: according to the recommended scenic spots, the number of buildings in the scenic spots, the number of deeds of people related to the buildings, the number of visitors and the duration of visits are obtained, the number of deeds of people related to the buildings, the number of visitors and the duration of visits are analyzed to obtain a building recommendation value, and a building recommendation judgment value is obtained from the building recommendation value and the number of buildings in the scenic spots, and a building with a building recommendation value greater than the building recommendation judgment value is recommended as a featured building;

[0295] The specific process of obtaining the building recommendation value is as follows:

[0296] Step S41: obtaining the number of buildings in the recommended scenic spot; the number of deeds of people related to the buildings; obtaining the proportion of deeds of people related to each building from the number of buildings in the recommended scenic spot and the number of deeds of people of each building, recorded as hc;

[0297] The proportion of the deeds of the characters related to each building is calculated as follows:

[0298]

[0299] Among them: jz is the number of buildings in the scenic spot, rwc is the number of deeds of people related to the buildings, <c≤jz;

[0300] Step S42: Obtain the number of visitors and the visiting time of each building in the past three months; obtain the total visiting time of tourists in each building from the number of visitors and the visiting time, recorded as zcc;

[0301] The total visit time of each building is calculated as follows:

[0302]

[0303] Among them: rsc is the number of visitors to each building, sccr is the length of time that visitors spend visiting each building;

[0304] Step S43: according to the total visit time of tourists, the number of visitors in each building in the previous month, and the visit time of tourists in each building, the visit time change value of the building is obtained, which is recorded as bhc;

[0305] The change in the building's visit duration is calculated as follows:

[0306]

[0307] Among them: dsc is the number of visitors to each building in the adjacent month, dccr is the visit time of tourists in each building in the adjacent month;

[0308] Step S44: Obtain the building recommendation value based on the total visit time of tourists, the proportion of building-related figures and deeds, and the change value of visit time, which is recorded as tjc;

[0309] The specific calculation of the building recommended value is as follows:

[0310] tjc=zcc×hc×bhc;

[0311] Step S45: Obtaining a building recommendation judgment value from the building recommendation value and the number of buildings in the scenic spot, denoted as pj;

[0312] The specific calculation process of the building recommendation judgment value is as follows:

[0313]

[0314] Step S46: Compare the building recommendation value with the building recommendation judgment value, and take the building whose recommendation value is greater than the building recommendation judgment value as the characteristic building of the scenic spot, and recommend the characteristic building to the user.

[0315] It should be noted that in the present application, if corresponding calculation formulas appear, the above calculation formulas are all dimensionless and take their numerical calculations. The weight coefficients, proportional coefficients and other coefficients in the formulas are set to a result value obtained by quantifying each parameter. The size of the weight coefficients and proportional coefficients can be determined as long as they do not affect the proportional relationship between the parameters and the result values.

[0316] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for recommending architectural features of tourist attractions based on big data analysis, characterized in that: The recommended methods include: Step S1: Obtain the user's geographical location, hobbies, and browsing history to obtain user information, and obtain the name of the scenic spot, scenic spot characteristics, scenic spot architectural style, scenic spot historical and cultural background, and scenic spot traffic information to obtain scenic spot information; Step S2: Processing scenic spot information and user information; counting scenic spot feature keywords according to scenic spot features in the scenic spot information, searching for scenic spot feature keywords through interests and browsing records in the user information, obtaining user preferences, combining user preferences with scenic spot information, obtaining correlation values ​​between scenic spots and user preferred scenic spots, and calculating correlation value deviations of scenic spots; The geographical location of the scenic spot is obtained according to the name of the scenic spot, and the relative distance value from the user to each scenic spot is obtained from the geographical location of the user and the geographical location of the scenic spot, and the weight value of the geographical location of the scenic spot is obtained from the relative distance value; The recommended weight value of the scenic spot is obtained by analyzing and calculating the architectural style, historical and cultural background, and visitor flow information of the scenic spot; Step S3: combining the correlation value deviation, the geographical location weight value and the recommendation weight value to obtain the recommended deviation of the scenic spot; obtaining the recommended scenic spot based on the recommended deviation; Step S4: According to the recommended scenic spots, the architectural information of the scenic spots is obtained, the architectural information of the scenic spots is analyzed to obtain the building recommendation value, the building recommendation judgment value is obtained from the building recommendation value and the building information, and the characteristic buildings are recommended according to the building recommendation value and the building recommendation judgment value.

2. The method for recommending architectural features of tourist attractions based on big data analysis according to claim 1, characterized in that: The specific steps of step S2 are as follows: Step S21: Obtaining scenic spot feature keywords through scenic spot features in the scenic spot information, and counting all scenic spot feature keywords; Step S22: taking the scenic spot information of all scenic spots as a set, denoted as D, and counting the number of scenic spots |D| in the set D; obtaining the number of scenic spot architectural styles according to the scenic spot information, and counting the number of scenic spots of each architectural style to obtain the number of scenic spots of each architectural style; Step S23: Obtaining a recommended value for a scenic spot based on the historical and cultural background of the scenic spot and the age group of the user; Step S24: retrieve the scenic spot feature keywords according to the interests and browsing records in the user information to obtain the user preferences, retrieve the user preferences, obtain the frequency of occurrence of the user preferences in the user information, and obtain the top sl user preferences with the highest frequency; put these preferences in the same set as the user's preference set; the number of preferences in the preference set is sl; Step S25: obtaining a correlation value between the scenic spot and the user preference according to the user preference and the scenic spot information, and calculating the correlation value deviation based on the correlation value; Step S26: obtaining the geographical location of the scenic spot according to the name of the scenic spot, and combining the geographical location of the user in the user information to obtain the relative distance |x|j between the geographical location of the user and the geographical location of the scenic spot; Calculate the ratio of the distance between each scenic spot and the user based on the relative distance, and obtain the geographical location weight value of the scenic spot, denoted as f(|x|j); The specific calculation process of the weight value of the scenic spot's geographical location is as follows: Where: |D| is the number of scenic spots, 0 <j≤|D|; Step S27: Obtain a first weight based on the tourist flow information of the scenic spot in the past three years; obtain a second weight by combining the first weight and the proportion of the number of scenic spots of each architectural style in the number of all scenic spots; obtain a recommended weight value by combining the second weight and the scenic spot recommendation value, and enter step S3.

3. The method for recommending architectural features of tourist attractions based on big data analysis according to claim 2, characterized in that: The specific steps of step S23 are as follows: Step S231: Obtain the number of users of different age groups who like b dynasties cd1, cd2, cd3, ... cdb through a questionnaire survey; Step S232: Calculate the proportion of the number of people who like each dynasty in the total number of people to obtain the proportion of the number of people who like each dynasty, denoted as zb; The specific calculation of the population ratio is as follows: Among them: cdx represents the number of users in the age group who like the xth dynasty, and the value range of x is: 1≤x≤b; Step S233: sort the population ratio values ​​in descending order; obtain the rank of each scenic spot in the ranking, and obtain the ranking value cw; obtain the historical and cultural background of the scenic spot according to the scenic spot information, obtain the historical figures related to each scenic spot in the historical and cultural background of the scenic spot, and obtain the number of historical figures gs associated with the scenic spot, and obtain the recommended value of the scenic spot in combination with the ranking value cw, which is recorded as tjz; The calculation steps for the recommended value of scenic spots are as follows:

4. The method for recommending architectural features of tourist attractions based on big data analysis according to claim 2, characterized in that: The specific steps of step S25 are as follows: Step S251: according to the scenic spot information, obtain the number T of all words in each scenic spot information; traverse the preference set, obtain the number of occurrences t of the user's preference in each scenic spot information, and combine the number T of all words in the scenic spot information to obtain the word frequency TF of the user's preference; The word frequency calculation process of user preference is as follows: Obtain the number of attractions that contain user preferences in all attraction information and calculate the inverse document frequency; The preferred inverse document frequency IDF calculation process is as follows; Where |D| is the number of scenic spots, and g(i) represents the number of scenic spots preferred by the i-th user in the preference set; Obtain the correlation value p between user preference and scenic spot based on the user's preferred word frequency TF and inverse document frequency IDF; The calculation process of the correlation value p between user preferences and attractions is as follows: p = TF × IDF; Count all the related values ​​and get the matrix A; Among them, p11 refers to the correlation value between the first user preference and the first scenic spot in the preference set, and so on, pij refers to the correlation value between the i-th user preference and the j-th scenic spot in the preference set; Step S252: Calculate the correlation value of each preference in the preference set to obtain the correlation value deviation of the scenic spot, recorded as Ej; The specific calculation steps of the correlation value deviation Ej are as follows: Where pij is the corresponding value in the matrix A, sl is the number of preferences in the preference set, 1≤i≤sl.

5. The method for recommending architectural features of tourist attractions based on big data analysis according to claim 2, characterized in that: The specific steps of step S27 are as follows: Step S271: according to the tourist flow information of the scenic spot, obtain the tourist flow num1, num2, num3 of the scenic spot in the past three years; obtain the first weight according to the tourist flow of the scenic spot in the past three years; Step S272: Obtain the number of scenic spots of each architectural style and the number of all scenic spots, calculate the proportion of the scenic spot architectural style, obtain the proportion value of the scenic spot architectural style, and obtain the second weight by combining the proportion value with the first weight; Step S273: combining the scenic spot recommendation value and the second weight to obtain a recommendation weight value, denoted as kk; The specific calculation process of the recommended weight value is as follows: kk=k2×tjz; Among them, k2 is the second weight, and tjz is the recommended value of the scenic spot.

6. A method for recommending architectural features of tourist attractions based on big data analysis according to claim 5, characterized in that: The specific steps of step S271 are as follows: Step S2711: Calculate the average number of visitors to the scenic spot based on the number of visitors to the scenic spot in the past three years, and record it as year(dj); The average visitor flow of an attraction is calculated as follows: Among them: dj represents a specific scenic spot; num1, num2, and num3 represent the annual passenger flow calculated one year, two years, and three years forward from the current year respectively; Step S2712: Obtain the average value of the scenic spots by combining the average value of the flow of people at the scenic spots with the number of scenic spots, recorded as all; The average value of the scenic spot is calculated as follows: Among them: year represents the average flow of tourists to the scenic spot, |D| is the number of scenic spots, 0 <j≤|D|; Step S2713: Obtain the first recommendation weight according to the average flow rate of the scenic spot, the average flow rate of the scenic spot, and the flow rates num1 and num3, denoted as k1: The specific calculation process of the first recommendation weight is as follows:

7. The method for recommending architectural features of tourist attractions based on big data analysis according to claim 5, characterized in that: The specific steps of step S272 are as follows: Step S2721: by calculating the number of scenic spots with this architectural style and the number of all scenic spots, a proportion of the scenic spot architectural style is obtained, which is recorded as bl; The calculation process of the proportion of scenic spot architectural style is as follows: Where: Gu is the number of scenic spots with the u-th architectural style, |D| is the number of all scenic spots; Step S2722: Calculate the proportion of the architectural style of the scenic spot and the first weight to obtain a second weight, recorded as k2; The specific calculation process of the second weight is as follows: k2=k1×bl; Among them: k1 is the first weight.

8. The method for recommending architectural features of tourist attractions based on big data analysis according to claim 1, characterized in that: The specific steps of step S3 are as follows: Step S31: Obtain the recommended deviation of the scenic spot according to the correlation value deviation, the geographical location weight value, and the recommended weight value, which is recorded as y; The recommended deviation is calculated as follows: Where: f(|x|j) is the geographical location weight value, Ej is the correlation value deviation, kk is the recommended weight value, 0 <j≤|D|; Step S32: sort the recommended deviations in descending order, and obtain the scenic spot with the smallest recommended deviation as the recommended scenic spot.

9. The method for recommending architectural features of tourist attractions based on big data analysis according to claim 1, characterized in that: The specific steps of step S4 are as follows: Step S41: Obtain the number of buildings in the recommended scenic spot, the number of deeds of people related to the buildings, the number of visitors and the duration of the visit to obtain building information, and obtain the proportion of deeds of people related to each building from the number of buildings in the recommended scenic spot and the number of deeds of people related to the buildings, recorded as hc; The proportion of building-related deeds is calculated as follows: Among them: jz is the number of buildings in the scenic spot, rwc is the number of deeds of people related to the buildings, <c≤jz; Step S42: according to the building information of the recommended scenic spot, the number of visitors and the visiting time of each building in the past three months are obtained; the total visiting time of tourists in each building is obtained from the number of visitors and the visiting time of each building, which is recorded as zcc; The total time spent by tourists visiting each building is calculated as follows: Among them: rsc is the number of visitors to each building in the past three months, sccr is the length of time that visitors spend visiting each building in the past three months; Step S43: according to the building information of the recommended scenic spot, the number of visitors and the visiting time of each building in the previous month are obtained; according to the total visiting time of tourists, the number of visitors and the visiting time of each building in the previous month, the visiting time change value of the building is obtained, which is recorded as bhc; The change in the building's visit duration is calculated as follows: Among them: dsc is the number of visitors to each building in the adjacent month, dccr is the visit time of tourists in each building in the adjacent month; Step S44: Obtain the building recommendation value based on the total visit time zcc of tourists, the proportion of building-related figures and deeds hc, and the visit time change value bhc, which is recorded as tjc; The specific calculation of the building recommended value is as follows: tjc=zcc×hc×bhc; Step S45: Obtain a building recommendation judgment value, denoted as pj, from the building recommendation value tjc and the number of buildings in the scenic spot jz; The specific calculation process of the building recommendation judgment value is as follows: Step S46: Compare the building recommendation value with the building recommendation judgment value, and take the building whose building recommendation value is greater than the building recommendation judgment value as the characteristic building of the scenic spot, and recommend the characteristic building to the user.

10. A device for recommending architectural features of tourist attractions based on big data analysis, applicable to a method for recommending architectural features of tourist attractions based on big data analysis as claimed in any one of claims 1 to 9, characterized in that: Recommended equipment includes: Collection module: used to obtain the user's geographic location, hobbies, and browsing history to obtain user information, and obtain the name of the scenic spot, the characteristics of the scenic spot, the architectural style of the scenic spot, the historical and cultural background of the scenic spot, and the flow of people in the scenic spot to obtain the scenic spot information; Data processing module: used to process scenic spot information and user information; obtain user preferences through user information, combine user preferences with scenic spot information, obtain the correlation value between scenic spots and user preferred scenic spots, and calculate the correlation value deviation of scenic spots; The geographical location of the scenic spot is obtained according to the name of the scenic spot, and the relative distance value from the user to each scenic spot is obtained from the geographical location of the user and the geographical location of the scenic spot, and the weight value of the geographical location of the scenic spot is obtained from the relative distance value; the recommended weight value of the scenic spot is obtained by analyzing and calculating the architectural style of the scenic spot, the historical and cultural background of the scenic spot, and the flow of people in the scenic spot; Calculation module: used to calculate the relevant value deviation, the weight value of the scenic spot's geographical location and the recommended weight value, obtain the recommended deviation of the scenic spot, and obtain the recommended scenic spot based on the recommended deviation; Recommendation module: According to the recommended scenic spots, the number of buildings in the scenic spots, the number of deeds of people related to the buildings, the number of visitors and the duration of visits are obtained. The number of deeds of people related to the buildings, the number of visitors and the duration of visits are analyzed to obtain the building recommendation value. According to the building recommendation value and the number of buildings in the scenic spots, the building recommendation judgment value is obtained. According to the building recommendation value and the building recommendation judgment value, characteristic buildings are recommended.

Images