Tourism information recommendation method and device based on big data

Abstract

The invention relates to the technical field of tourism information, in particular to a tourism information recommendation method and device based on big data. The method comprises the following steps: respectively obtaining a user travel intention label and a scenic spot identification label, and outputting a recommendation place list in a multi-layer matching manner; acquiring historical people flow of the recommended place, forming time sequence data, setting an early warning threshold value, performing current people flow early warning and prediction according to the real-time people flow, and outputting the current people flow early warning and prediction; dynamically adjusting a recommended place list according to current pedestrian flow early warning and prediction and in combination with current activity data of the user; the device comprises a recommended place matching module, an early warning and predicting module and a recommended place dynamic adjusting module. According to user preferences, label matching is performed in a targeted manner, recommendation places are acquired layer by layer, current visitor flow early warning is performed through historical visitor flow and real-time visitor flow, prediction data and recommendation data are dynamically output, tourism information recommendation accuracy is improved, and tourist experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of tourism information technology, and in particular to a tourism information recommendation method and device based on big data. Background Art

[0002] With the continuous expansion of my country's tourism industry and the improvement of people's living standards, more and more people are choosing to travel to improve their quality of life. People can travel by joining a group tour, with the itinerary planned by a travel agency, but this can lead to significant limitations. To avoid these limitations, independent travel offers more freedom, but requires people to plan their own destinations using relevant software.

[0003] However, current related software usually determines popular tourist spots based on the popularity of scenic spots and historical tour data when recommending travel information. Popular tourist spots may be more popular during a certain period of time and have a large flow of people, thereby reducing the user experience at the tourist spot. Summary of the Invention

[0004] The purpose of the present invention is to provide a tourism information recommendation method and device based on big data, aiming to solve the technical problem in the prior art that in tourism information recommendation, popular tourist spots are usually determined based on the popularity of the scenic area and historical tour data. Popular tourist spots may be more popular in a certain period of time and have a large flow of people, thereby reducing the user experience of the tourist at the tourist spot.

[0005] To achieve the above-mentioned purpose, the present invention adopts a tourism information recommendation method based on big data, which includes the following steps: Obtain user travel intention labels and tourist attraction identification labels respectively, and output a list of recommended places through multi-layer matching; Obtain historical pedestrian flow at the recommended location, generate time series data, set warning thresholds, and generate and output current pedestrian flow warnings and forecasts based on real-time pedestrian flow. The recommended location list is dynamically adjusted based on current crowd flow warnings and forecasts, combined with the user's current activity data.

[0006] Among them, in the steps of respectively obtaining the user's travel intention tag and the tourist attraction identification tag, and outputting the recommended place list through multi-layer matching, the process of obtaining the tourist attraction identification tag is as follows: Collect information about tourist attractions and label the tourist attractions based on the information; the information about tourist attractions includes introduction to the attractions, description of features, visitor reviews, geographical location, and surrounding facilities.

[0007] Among them, in the steps of respectively obtaining the user's travel intention label and the tourist attraction identification label, and outputting the recommended destination list through multi-layer matching, the process of obtaining the user's travel intention label is as follows: Collect basic user data, perform text mining and analysis on the basic data, extract basic user tags, compare the basic user tags with the tourist attraction identification tags, and output preliminary recommended destinations; the basic data includes destination, travel type, and travel time preference; Collect user behavior data, perform text mining and analysis on the behavior data, extract user behavior tags, compare the user behavior tags with the identification tags of the preliminary recommended places, and output in-depth recommended places; the behavior data includes browsing, booking, and review behaviors; Collect user demand data, conduct text mining and analysis on the demand data, extract user demand labels, compare the user demand labels with the identification labels of deeply recommended places, output personalized recommended places, and form a list of recommended places; among them, the demand data includes the number of people traveling, the eating habits of multiple people, and the interaction situation.

[0008] Among them, in the steps of text mining and analysis: Preprocess the acquired data and output segmented data; preprocessing includes data cleaning, text segmentation, part-of-speech tagging, and stop word removal; Count the frequency of each word in the word segmentation data and extract features based on the importance of each word; Get multiple feature combinations to generate labels.

[0009] Among them, in the steps of obtaining the historical pedestrian flow of the recommended location, forming time series data, setting the warning threshold, and making current pedestrian flow warning and prediction based on the real-time pedestrian flow, and outputting it: Obtain historical traffic data of the recommended location and organize the historical traffic data in chronological order to form time series data; Analyze the distribution of historical pedestrian flow data, calculate the average pedestrian flow and standard deviation, and set warning thresholds based on the calculation results; the warning thresholds include the third warning threshold, the second warning threshold, and the first warning threshold.

[0010] After analyzing the distribution of historical pedestrian flow data, calculating the average pedestrian flow and standard deviation, and setting the third warning threshold, the second warning threshold, and the first warning threshold based on the calculation results: Collect real-time pedestrian flow at the recommended location, obtain pedestrian flow data, compare the pedestrian flow data with the warning threshold, and output the current warning data.

[0011] After collecting the real-time pedestrian flow at the recommended location, obtaining pedestrian flow data, comparing the pedestrian flow data with the warning threshold, and outputting the current warning data: Make crowd flow predictions based on historical crowd flow data and real-time crowd flow.

[0012] Among them, in the step of dynamically adjusting the recommended location list based on the current pedestrian flow warning and prediction, and combining the user's current activity data: Obtain the user's current activity data and current crowd flow warning and prediction results; the user's current activity data includes current location data and travel data; Adjust the list of recommended places based on crowd flow warnings and predictions.

[0013] Among them, after the step of adjusting the recommended location list based on crowd flow warning and prediction: Adjust the recommended list based on the user's current activity data.

[0014] The present invention also provides a tourism information recommendation device based on big data, comprising a recommended location matching module, an early warning and prediction module, and a recommended location dynamic adjustment module; wherein: The recommended place matching module is used to obtain the user's travel intention tag and the tourist attraction identification tag respectively, and output a recommended place list through multi-layer matching; The warning and prediction module is used to obtain the historical pedestrian flow of the recommended location, form time series data, set warning thresholds, and perform current pedestrian flow warning and prediction based on real-time pedestrian flow, and output the results; The recommended places dynamic adjustment module is used to dynamically adjust the recommended places list according to the current pedestrian flow warning and prediction and in combination with the user's current activity data.

[0015] The present invention provides a method and device for recommending tourism information based on big data, which respectively adopts the recommended place matching module, the early warning and prediction module, and the recommended place dynamic adjustment module to perform the following steps: respectively obtain the user's travel intention label and the tourist attraction identification label, and output a recommended place list through multi-layer matching; obtain the historical flow of people in the recommended places, form time series data, set an early warning threshold, and perform current flow early warning and prediction based on the real-time flow, and output; dynamically adjust the recommended place list based on the current flow early warning and prediction and in combination with the user's current activity data; through targeted label matching based on the user's preferences, obtain recommended places layer by layer, perform current flow early warning based on historical flow and real-time flow, dynamically output prediction data and recommendation data, improve the accuracy of tourism information recommendation, and thus enhance the tourist experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 It is a flow chart of the tourism information recommendation method based on big data of the present invention.

[0018] Figure 2 It is a flowchart of the steps of the tourism information recommendation method based on big data of the present invention.

[0019] Figure 3 It is a step flow chart of S100 of the present invention.

[0020] Figure 4 It is a step flow chart of S200 of the present invention.

[0021] Figure 5 It is a step flow chart of S300 of the present invention.

[0022] Figure 6 It is a structural principle diagram of the tourism information recommendation device based on big data of the present invention.

[0023] Figure 7 It is a structural principle diagram of the electronic device of the present invention.

[0024] 401-Recommended location matching module, 402-Warning and prediction module, 403-Recommended location dynamic adjustment module. DETAILED DESCRIPTION

[0025] Exemplary embodiments are described in detail herein, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, identical numerals in different drawings represent identical or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with this application.

[0026] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this application and the appended claims, the singular forms "a," "an," "the," and "the" are intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0027] It should be understood that although the terms first, second, third, etc. may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0028] See also Figures 1 to 5 The present invention provides a method for recommending tourism information based on big data, comprising the following steps: S100: Obtain user travel intention tags and tourist attraction identification tags respectively, and perform multi-layer matching to output a list of recommended places.

[0029] In this embodiment, the user's travel intention tags and tourist attraction identification tags are obtained separately, and a multi-layer matching is performed to output a list of recommended places. The specific process is as follows: S101: Collecting tourist attraction information and labeling the tourist attraction according to the tourist attraction information; wherein the tourist attraction information includes an introduction to the attraction, a description of its characteristics, visitor reviews, its geographical location, and surrounding facilities; S102: Collect user basic data, perform text mining and analysis on the basic data, extract user basic tags, compare the user basic tags with tourist attraction identification tags, and output preliminary recommended destinations; the basic data includes destination, travel type, and travel time preference; S103: Collect user behavior data, perform text mining and analysis on the behavior data, extract user behavior tags, compare the user behavior tags with the identification tags of the preliminary recommended places, and output in-depth recommended places; the behavior data includes browsing, booking, and review behaviors; S104: Collect user demand data, perform text mining and analysis on the demand data, extract user demand labels, compare the user demand labels with the identification labels of the deeply recommended places, output personalized recommended places, and form a list of recommended places; among them, the demand data includes the number of people traveling, the eating habits of multiple people, and the interaction situation.

[0030] The process of text mining and analysis is as follows: Preprocess the acquired data and output segmented data; preprocessing includes data cleaning, text segmentation, part-of-speech tagging, and stop word removal; Count the frequency of each word in the word segmentation data and extract features based on the importance of each word; Get multiple feature combinations to generate labels.

[0031] In the above process, both user tags and scenic spot tags can be obtained by text mining and analysis: the obtained data is cleaned to remove irrelevant characters, punctuation marks, special symbols, etc.

[0032] Use word segmentation tools to split text data into individual words, mark the part of speech of each word after segmentation, such as noun, verb, adjective, etc., and remove some common words that have no practical meaning, such as stop words.

[0033] Count the frequency of each word in the text and use high-frequency words as candidate features. Use the TF-IDF (Term Frequency-Inverse Document Frequency) algorithm to calculate the importance of each word and further select representative features. The calculation process of this algorithm is as follows: Get the term frequency (TF); where term frequency represents the frequency of a term in a document. For a term t in document d, the term frequency TF(t,d) is calculated as: Where: the number of occurrences of word t is the number of times word t appears in document d; The total number of all words in document d is the total number of all words in document d.

[0034] Get the Inverse Document Frequency (IDF), which measures the importance of a word in the entire document collection. For a word t, the inverse document frequency IDF(t) is calculated as: Where: N is the total number of documents in the document collection; the number of documents containing word t is the number of documents in the document collection that contain word t.

[0035] Calculate the TF-IDF value; the TF-IDF value is the product of the term frequency and the inverse document frequency, which is used to indicate the importance and discrimination of a word in a document. For a word t in document d, the TF-IDF value w t,d , the calculation formula is: TF-IDF value w t,d The higher the value, the more frequently the word appears in the document, but less frequently in other documents. Therefore, the word is more important and discriminative to the document.

[0036] TF-IDF value w t,d The lower the value, the less frequent the word appears in the document, or the more frequent the word appears in other documents. Therefore, the less important and discriminative the word is to the document.

[0037] Based on the above tag acquisition method, user tags and scenic spot tags are acquired respectively; wherein: Tourist attractions are labeled with information such as attraction introduction, feature description, visitor reviews, geographical location, and surrounding facilities.

[0038] Based on the user's destination, travel type, and travel time preference data, such as the user's frequent searches for keywords such as "ancient town tourism" and "cultural tour," we extract the user's basic tags, compare them with the tourist attraction identification tags, and output preliminary recommendations. The following matching method is used to compare the user's basic tags with the tourist attraction identification tags: Assume that the user basic label vector and the tourist attraction identification label vector are: User basic label vector: U=(u1,u2,…,u n ); Tourist attraction identification label vector: P=(p1,p2,…,p n ); Among them, u i and p i Respectively represent the weights of users and attractions on the i-th tag, which can be TF-IDF values; The calculation formula of cosine similarity sim(U,P) is: By calculating the cosine similarity between the user base tag vector and the tourist attraction tag vector, the degree of matching can be obtained. The closer the similarity value is to 1, the higher the degree of matching.

[0039] Based on the user's browsing, booking, and evaluation behavior data, such as users frequently browsing pages of high-end resort hotels, user behavior tags are extracted. Using the above matching method, the user behavior tags are compared with the identification tags of the preliminary recommended places to output the deeply recommended places.

[0040] Based on the number of users traveling, their diet, and their interactions, such as whether there are elderly people or children traveling, user demand labels are extracted. Using the above matching method, the user demand labels are compared with the identification labels of the deeply recommended places, and personalized recommended places are output to form a list of recommended places.

[0041] In the multi-layer matching process, a similarity threshold θ can be set. Only when the cosine similarity between the user's demand label and the identification label of the deep recommended place is greater than or equal to θ, the attraction will be included in the recommended place list.

[0042] in: Recommended places list = {attractions P | sim(U,P) ≥ θ} By adjusting the threshold θ, the length and quality of the recommendation list can be controlled. A higher threshold will generate a more accurate but smaller recommendation list, while a lower threshold will generate more but potentially less relevant recommendation lists.

[0043] S200: Obtain historical pedestrian flow at the recommended location, generate time series data, set warning thresholds, and perform current pedestrian flow warning and prediction based on real-time pedestrian flow, and output the data.

[0044] In this implementation, the historical flow of people at the recommended location is obtained to form time series data, an early warning threshold is set, and the current flow of people is warned and predicted based on the real-time flow of people, and then output. The specific process is as follows: S201: Obtain historical pedestrian flow data of the recommended location, and organize the historical pedestrian flow data in chronological order to form time series data; S202: Analyze the distribution of historical pedestrian flow data, calculate the average pedestrian flow and standard deviation, and set warning thresholds based on the calculation results; the warning thresholds include a third warning threshold, a second warning threshold, and a first warning threshold; S203: Collecting real-time pedestrian flow at the recommended location and obtaining pedestrian flow data, comparing the pedestrian flow data with the warning threshold, and outputting current warning data; S204: Perform crowd flow prediction based on historical crowd flow data and real-time crowd flow.

[0045] In the above process, the historical flow data of the recommended place in the past period (such as one year or six months) is obtained from the management department of the recommended place, third-party data platforms (such as map navigation software), etc. The data granularity can be daily, hourly or even shorter time intervals. The collected historical flow data is sorted in chronological order to form time series data X={x1,x2,⋯,x t}, where x t represents the flow of people at the tth time point.

[0046] Analyze the distribution of historical traffic data, such as calculating the average traffic flow and standard deviation σ; Average passenger flow The calculation method is: The standard deviation σ is calculated as: Set the warning threshold, for example, the yellow warning is the third warning threshold +σ, orange warning is the second warning threshold +1.5σ, red warning is the first warning threshold +2σ.

[0047] Based on historical traffic data and real-time traffic, traffic flow is predicted. The prediction model uses the SARIMA (Seasonal Autoregressive Moving Average) model in the time series model. The formula is: in, is the backshift operator, For seasonal cycles, and are seasonal autoregressive and moving average polynomials, respectively. and are non-seasonal autoregressive and moving average polynomials, respectively. and are the non-seasonal and seasonal differencing orders, respectively, is the white noise error term. The SARIMA model is trained and parameterized using historical pedestrian flow time series data. The trained SARIMA model is used to predict pedestrian flow over a period of time and output the predicted results.

[0048] S300: Dynamically adjust the recommended location list based on current pedestrian flow warnings and predictions, and in combination with the user's current activity data.

[0049] In this embodiment, the recommended places list is dynamically adjusted based on the current pedestrian flow warning and prediction, combined with the user's current activity data. The specific process is as follows: S301: Obtaining the user's current activity data and current crowd flow warning and prediction results; wherein the user's current activity data includes current location data and travel data; S302: Adjust the recommended location list based on pedestrian flow warning and prediction; S303: Adjust the recommended location list based on the user's current activity data.

[0050] In the above process, the user's current activity data and the current crowd flow warning and prediction results are obtained; wherein the user's current activity data includes current positioning data and travel data.

[0051] Adjust the recommended destination list based on crowd flow warnings and forecasts: If a recommended destination currently triggers a red alert and crowd flow is forecast to remain high for a period of time, the attraction will be removed from the recommended destination list or its ranking will be lowered; if a destination currently has low crowd flow and the crowd flow forecast is also low for a period of time, its ranking in the recommended destination list will be raised; Adjust the recommended destination list based on the user's current activity data: Consider the user's current location and itinerary. If the user is already near a certain attraction and the attraction currently has a large flow of people, you can recommend some relatively unpopular attractions with similar characteristics in the surrounding area; if the user's itinerary already includes a popular attraction, you can avoid recommending the attraction repeatedly in the recommended destination list and instead recommend some complementary attractions.

[0052] The recommended list is adjusted as shown in the following example: For each attraction in the recommended list, view its current crowd warning level and future crowd forecast. For example, for five attractions (A, B, C, D, and E) in the recommended list, obtain the warning level and forecast trend for each of the five attractions.

[0053] If a recommended destination currently triggers a red alert and the number of visitors is predicted to remain high for a period of time, the attraction will be removed from the recommended destination list or its ranking will be lowered. For example, if attraction A currently triggers a red alert and the number of visitors is predicted to remain high for the next three hours, it will be removed from the recommended destination list. If the attraction is well-known, its ranking can also be lowered from 1st to 5th.

[0054] If a particular attraction currently has low traffic and is forecast to have low traffic for the next few hours, its ranking in the recommended destination list will be raised. For example, if attraction B currently has a green traffic alert and is forecast to have low traffic for the next three hours, its ranking will be raised from 4th to 2nd.

[0055] Based on the user's current location, the distance between the user and each attraction in the recommended list is determined to determine whether the user is near a particular attraction. For example, if the user's current location shows that they are only 500 meters away from attraction C, it can be determined that the user is near attraction C.

[0056] Combined with the user's itinerary, the system can understand the attractions the user has planned to visit and the travel time that has not yet been scheduled. For example, the user's itinerary shows that they have planned to visit attraction D in the afternoon of that day, but the afternoon schedule is relatively tight, with only 2 hours of free time remaining.

[0057] Adjust the recommended destination list to recommend nearby less popular, similar attractions: If the user is already near a popular attraction with high traffic, recommend nearby less popular attractions with similar characteristics. For example, if the user is already near attraction C (a popular attraction with high traffic), and analysis of surrounding attraction information reveals that attraction F has a similar theme to C (e.g., both are historical and cultural attractions) but has lower traffic, then add attraction F to the recommended destination list and give it a higher ranking.

[0058] Avoid duplicate recommendations and recommend complementary attractions: If a popular attraction is already on a user's itinerary, avoid recommending it again in the recommended destination list and instead recommend complementary attractions. For example, if a user's itinerary already includes attraction D (focused on natural scenery), when adjusting the recommended destination list, instead of recommending a similar natural attraction like attraction D, we will recommend attraction G (focused on folk culture experiences) to complement D and enrich the user's travel experience.

[0059] After the two phases of adjustments described above, a final list of recommended destinations is generated, combining crowd flow warnings and forecasts with the user's current activity data. This list is presented to the user in adjusted ranking order, ensuring travel recommendations that better suit their actual circumstances and needs. For example, the final list of recommended destinations might be: Attraction B (ranked 1st), Attraction F (ranked 2nd), Attraction G (ranked 3rd), Attraction E (ranked 4th), and Attraction D (although popular, it is already included in the itinerary, ranked lower, or not recommended).

[0060] In this invention, user travel intention tags and tourist attraction identification tags are obtained separately, and a multi-layer matching process is used to output a list of recommended destinations. The historical footfall of recommended destinations is then collected to form time series data, and warning thresholds are set. Current footfall warnings and predictions are generated based on real-time footfall, and these are then output. Finally, the list of recommended destinations is dynamically adjusted based on current footfall warnings and predictions, combined with the user's current activity data. By performing targeted tag matching based on user preferences, obtaining recommended destinations layer by layer, and using historical and real-time footfall warnings to generate current footfall warnings, and dynamically outputting predicted and recommended data, the accuracy of travel information recommendations is improved, thereby enhancing the visitor experience.

[0061] Corresponding to the aforementioned embodiment of the tourism information recommendation method based on big data, the present application also provides an embodiment of the tourism information recommendation device based on big data.

[0062] Figure 6 This is a block diagram of a travel information recommendation device based on big data according to an exemplary embodiment. Figure 6 The device may include: a recommended location matching module 401, an early warning and prediction module 402, and a recommended location dynamic adjustment module 403; wherein: The recommended destination matching module 401 is used to obtain the user's travel intention tag and the tourist attraction identification tag, perform multi-layer matching and output a recommended destination list; The warning and prediction module 402 is used to obtain historical pedestrian flow at the recommended location, generate time series data, set warning thresholds, and perform current pedestrian flow warning and prediction based on real-time pedestrian flow, and output the results; The recommended places dynamic adjustment module 403 is used to dynamically adjust the recommended places list according to the current pedestrian flow warning and prediction and in combination with the user's current activity data.

[0063] In this embodiment, the recommended place matching module 401 obtains the user's travel intention label and tourist attraction identification label respectively, and outputs a list of recommended places through multi-layer matching; the warning and prediction module 402 obtains the historical flow of people at the recommended places, forms time series data, sets warning thresholds, and performs current flow warning and prediction based on real-time flow, and outputs them; the recommended place dynamic adjustment module 403 dynamically adjusts the recommended place list based on the current flow warning and prediction, and in combination with the user's current activity data; by performing targeted label matching according to the user's preferences, recommended places are obtained layer by layer, and current flow warning is performed through historical flow and real-time flow, and prediction data and recommendation data are dynamically output to improve the accuracy of tourism information recommendation, thereby enhancing the tourist experience.

[0064] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0065] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0066] Accordingly, the present application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned method for recommending travel information based on big data. Figure 7 As shown in FIG, a hardware structure diagram of a device having data processing capability for recommending tourism information based on big data provided by an embodiment of the present invention is shown, except for Figure 7 In addition to the processor, memory, and network interface shown, any device with data processing capabilities in which the apparatus in the embodiment is located may also include other hardware, generally based on the actual functions of the device with data processing capabilities, which will not be described in detail.

[0067] Accordingly, the present application also provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the above-mentioned method for recommending travel information based on big data. The computer-readable storage medium may be an internal storage unit of any device with data processing capabilities described in any of the aforementioned embodiments, such as a hard disk or memory. The computer-readable storage medium may also be an external storage device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), an SD card, a flash card, etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capabilities. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capabilities, and may also be used to temporarily store data that has been output or is to be output.

[0068] Those skilled in the art will readily conceive of other embodiments of the present application after considering the specification and practicing the contents disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of this application and include common knowledge or customary techniques in the art that are not disclosed in this application.

[0069] It will be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A tourism information recommendation method based on big data, characterized in that: The steps include: Obtain user travel intention labels and tourist attraction identification labels respectively, and output a list of recommended places through multi-layer matching; Obtain historical pedestrian flow at the recommended location, generate time series data, set warning thresholds, and generate and output current pedestrian flow warnings and forecasts based on real-time pedestrian flow. The recommended location list is dynamically adjusted based on current crowd flow warnings and forecasts, combined with the user's current activity data.

2. The method for recommending tourism information based on big data according to claim 1, wherein: In the steps of respectively obtaining the user's travel intention tag and the tourist attraction identification tag, and outputting the recommended destination list through multi-layer matching, the process of obtaining the tourist attraction identification tag is as follows: Collect information about tourist attractions and label the tourist attractions based on the information; the information about tourist attractions includes introduction to the attractions, description of features, visitor reviews, geographical location, and surrounding facilities.

3. The method for recommending tourism information based on big data according to claim 2, wherein: In the steps of obtaining the user's travel intention label and the tourist attraction identification label, and outputting the recommended destination list through multi-layer matching, the process of obtaining the user's travel intention label is as follows: Collect basic user data, perform text mining and analysis on the basic data, extract basic user tags, compare the basic user tags with the tourist attraction identification tags, and output preliminary recommended destinations; the basic data includes destination, travel type, and travel time preference; Collect user behavior data, perform text mining and analysis on the behavior data, extract user behavior tags, compare the user behavior tags with the identification tags of the preliminary recommended places, and output in-depth recommended places; the behavior data includes browsing, booking, and review behaviors; Collect user demand data, conduct text mining and analysis on the demand data, extract user demand labels, compare the user demand labels with the identification labels of deeply recommended places, output personalized recommended places, and form a list of recommended places; among them, the demand data includes the number of people traveling, the eating habits of multiple people, and the interaction situation.

4. The method for recommending tourism information based on big data according to claim 3, wherein: In the steps of text mining and analysis: Preprocess the acquired data and output segmented data; preprocessing includes data cleaning, text segmentation, part-of-speech tagging, and stop word removal; Count the frequency of each word in the word segmentation data and extract features based on the importance of each word; Get multiple feature combinations to generate labels.

5. The method for recommending tourism information based on big data according to claim 1, wherein: In the steps of obtaining the historical pedestrian flow of the recommended location, forming time series data, setting warning thresholds, and making current pedestrian flow warnings and predictions based on real-time pedestrian flow, and outputting the data: Obtain historical traffic data of the recommended location and organize the historical traffic data in chronological order to form time series data; Analyze the distribution of historical traffic data, calculate the average traffic flow and standard deviation, and set warning thresholds based on the calculation results; The warning thresholds include a third warning threshold, a second warning threshold, and a first warning threshold.

6. The method for recommending tourism information based on big data according to claim 5, characterized in that: After analyzing the distribution of historical pedestrian flow data, calculating the average pedestrian flow and standard deviation, and setting the third warning threshold, the second warning threshold, and the first warning threshold based on the calculation results: Collect real-time pedestrian flow at the recommended location, obtain pedestrian flow data, compare the pedestrian flow data with the warning threshold, and output the current warning data.

7. The method for recommending tourism information based on big data according to claim 6, wherein: After collecting the real-time pedestrian flow at the recommended location, obtaining the pedestrian flow data, comparing the pedestrian flow data with the warning threshold, and outputting the current warning data: Make crowd flow predictions based on historical crowd flow data and real-time crowd flow.

8. The method for recommending tourism information based on big data according to claim 1, wherein: In the step of dynamically adjusting the recommended location list based on current crowd flow warnings and predictions, combined with the user's current activity data: Obtain the user's current activity data and current crowd flow warning and prediction results; the user's current activity data includes current location data and travel data; Adjust the list of recommended places based on crowd flow warnings and predictions.

9. The method for recommending tourism information based on big data according to claim 8, wherein: After adjusting the recommended location list based on crowd flow warnings and predictions: Adjust the recommended list based on the user's current activity data.

10. A tourism information recommendation device based on big data, applied to the tourism information recommendation method based on big data according to claim 1, characterized in that: It includes the recommended location matching module, the early warning and prediction module, and the recommended location dynamic adjustment module; among which: The recommended place matching module is used to obtain the user's travel intention tag and the tourist attraction identification tag respectively, and output a recommended place list through multi-layer matching; The warning and prediction module is used to obtain the historical pedestrian flow of the recommended location, form time series data, set warning thresholds, and perform current pedestrian flow warning and prediction based on real-time pedestrian flow, and output the results; The recommended places dynamic adjustment module is used to dynamically adjust the recommended places list according to the current pedestrian flow warning and prediction and in combination with the user's current activity data.

Images