CN117611395A - Intelligent travel decision-making method and system based on big data - Google Patents

Abstract

The invention provides a big data-based intelligent travel decision method and a big data-based intelligent travel decision system, wherein the method comprises the following steps: step 1: acquiring a travel plan which is input by a user and needs to be subjected to travel decision; step 2: acquiring tourist big data, and making a tourist guide suitable for a tourist plan based on the tourist big data; step 3: pushing the tour guide to the user as a tour decision result. According to the intelligent travel decision method and system based on big data, based on the big travel data, the proper travel guide is formulated according to the travel plan of the user, so that the user is helped to make reasonable travel decisions, convenience is improved, and meanwhile, the travel experience of the user is improved.

Description

Intelligent travel decision-making method and system based on big data

Technical Field

The invention relates to the technical field of big data, in particular to an intelligent travel decision method and system based on big data.

Background

At present, when a user needs to travel, a travel plan (such as which people travel, which places to go, playing for a few days and the like) is simply established, and then a detailed travel route (such as a travel attack of which places to go through Internet searching is established based on the travel attack) is established, so that the travel plan is complicated, the travel route established by the user is not necessarily reasonable, and travel experience is reduced;

Thus, a solution is needed.

Disclosure of Invention

The invention provides a big data-based intelligent travel decision method and a big data-based intelligent travel decision system, which are used for making a proper travel guide according to a user's travel plan based on the big travel data, helping the user to make a reasonable travel decision, improving convenience and improving the user's travel experience.

The invention provides an intelligent travel decision method based on big data, which comprises the following steps:

step 1: acquiring a travel plan which is input by a user and needs to be subjected to travel decision;

step 2: acquiring tourist big data, and making a tourist guide suitable for the tourist plan based on the tourist big data;

step 3: pushing the tour guide to a user as a tour decision result.

Preferably, in the step 2, the obtaining travel big data includes:

obtaining a preset travel big data source set, wherein the travel big data source set comprises: a plurality of first sources;

obtaining a credit value corresponding to the first source;

if the credit value is greater than or equal to a preset credit threshold, the corresponding first source is used as a second source;

obtaining a source type corresponding to the second source, wherein the source type comprises: individuals and platforms;

When the source type corresponding to the second source is personal, acquiring a first data attribute of at least one first data item provided by the second source, and simultaneously acquiring information to be verified, provided by the second source, corresponding to the first data item;

acquiring at least one first verification policy corresponding to the first data attribute, and verifying the corresponding information to be verified based on the first verification policy;

when the first data item passes the verification, the corresponding first data item is taken as a second data item;

when the source type corresponding to the second source is a platform, acquiring a second data attribute of at least one third data item provided by the second source, and simultaneously acquiring a guarantee value corresponding to the second source for guaranteeing the third data item;

acquiring a guarantee threshold corresponding to the second data attribute, and taking the corresponding third data item as a fourth data item if the guarantee value is greater than or equal to the guarantee threshold;

and integrating the second data item and the fourth data item to obtain travel big data, and completing the acquisition.

Preferably, obtaining the credit value corresponding to the first source includes:

Acquiring a plurality of credit records corresponding to the first source;

extracting the record characteristics of the credit records to obtain a plurality of record characteristics;

constructing a record characteristic-dynamic malicious value library, and determining a first dynamic malicious value corresponding to the record characteristic based on the record characteristic-dynamic malicious value library;

accumulating and calculating the first dynamic malicious value to obtain a malicious value sum;

acquiring the malicious value and a corresponding down-regulation strategy;

and acquiring a credit initial value corresponding to the first source, and performing down-regulation on the credit initial value based on the down-regulation strategy to acquire the credit value corresponding to the first source so as to finish the acquisition.

Preferably, constructing a record feature-dynamic malicious value library includes:

acquiring a preset bad credit record characteristic set, wherein the record characteristic set comprises: a plurality of second recorded features;

acquiring a result event corresponding to the second record feature, and simultaneously acquiring a feature weight of the second record feature corresponding to the result event;

based on a preset result severity judging model, carrying out result severity judgment on the result event to obtain a first severity value, and endowing the first severity value with the characteristic weight to obtain a to-be-counted value;

Accumulating and calculating the to-be-counted value to obtain a to-be-paired value;

combining and pairing the value to be paired with the corresponding second record characteristic to obtain a pairing item;

acquiring a preset blank library, and storing the pairing items into the blank library;

and after the pairing items which are required to be stored in the blank library are stored, the blank library is used as a record characteristic-dynamic malicious value library, and the construction is completed.

Preferably, in the step 2, based on the travel big data, a travel guide suitable for the travel plan is formulated, including:

training a travel guide to make a model;

and (3) setting a model based on the travel guide, and setting a travel guide suitable for the travel plan according to the travel big data.

Preferably, training the travel guideline to model includes:

acquiring a plurality of first travel guide making events;

acquiring an experience value of a formulator corresponding to the first travel guide formulation event, and taking the corresponding first travel guide formulation event as a second travel guide formulation event if the experience value is greater than or equal to a preset experience threshold;

acquiring a formulation process corresponding to the second travel guide formulation event;

splitting the formulation process into a plurality of first sub-processes;

Sequencing the first sub-process according to a preset process sequencing rule to obtain a first process sequence;

extracting process characteristics of the first sub-process to obtain a plurality of first process characteristics;

acquiring a preset verification trigger feature library, matching the first process features with second process features in the verification trigger feature library, taking the corresponding first subprocess as a second subprocess if the matching is met, and simultaneously acquiring at least one second verification strategy corresponding to the matched second process features;

verifying the second sub-process based on the second verification policy, and taking the corresponding second verification policy as a third verification policy if the verification is not passed;

at least one influence item corresponding to the third verification policy is obtained, and meanwhile, an influence type corresponding to the influence item is obtained, wherein the influence type comprises: internal and external influences;

when the influence type of the influence item is an internal influence, a first influence value corresponding to the influence item is obtained, and the first influence value is marked on the second subprocess;

when the influence type of the influence item is external influence, acquiring a second influence value, an influence direction, an influence range and an influence target corresponding to the influence item;

Determining the first sub-process corresponding to the influence target in the influence range in the influence direction of the second sub-process in the first process sequence, and taking the first sub-process as a third sub-process;

marking the second influence value on the third sub-process;

when the first influence value and the second influence value are marked, the first process sequence is used as a second process sequence;

accumulating the marked first influence value and the marked second influence value of each fourth sub-process in the second process sequence to obtain a sum of influence values;

acquiring the influence value and a corresponding process weight threshold value, and simultaneously acquiring a process weight of the fourth sub-process corresponding to the formulation process;

if the process weight is greater than or equal to the corresponding process weight threshold, eliminating the corresponding second travel guide making event;

when all the second travel guide making events needing to be removed are removed, taking the removed remaining second travel guide making events as third travel guide making events;

and performing model training on the third travel guide making event based on a preset model training algorithm to obtain a travel guide making model.

Preferably, the intelligent travel decision method based on big data further comprises:

extracting a travel route from the travel guide, wherein the travel route passes through at least one first scenic spot;

determining at least one first emergency corresponding to the first scenic spot from the tourist big data;

based on a preset severity judging model, carrying out severity judgment on the first sudden event, obtaining a second severity value, and associating with the corresponding first scenic spot;

accumulating and calculating the second serious value associated with the first scenic spot to obtain a serious value sum;

if the serious value sum is greater than or equal to a preset serious value sum and a threshold value sum, taking the corresponding first scenic spot as a second scenic spot, and taking the first emergency corresponding to the second scenic spot as a second emergency;

acquiring at least one first state behavior corresponding to the second emergency;

extracting at least one first travel person corresponding to the second scenic spot from the travel guide;

when the first traveler enters the second Jing Oushi, taking the first traveler entering the second scenic spot as a second traveler;

controlling a tour guide trolley in the second scenic spot to follow a tour team consisting of the second tour personnel;

In the following process, controlling the tour guide trolley to dynamically acquire a plurality of second state behaviors generated by the second trip personnel;

matching the second state behavior with the first state behavior, if the matching is met, acquiring the characterization degree corresponding to the first state behavior which is met by the matching, and correlating with the corresponding second emergency;

accumulating and calculating the characterization degree associated with the second emergency event to obtain a characterization degree sum;

acquiring a characterization degree and a threshold value corresponding to the second emergency, and taking the corresponding second emergency as a third emergency and taking the corresponding second trip personnel as a third trip personnel if the characterization degree and the threshold value are larger than or equal to each other;

at least one coping strategy corresponding to the third emergency is obtained, and meanwhile, a coping type corresponding to the coping strategy is obtained, wherein the coping type comprises: manual emergency handling and non-manual emergency handling;

when the coping type corresponding to the coping strategy is artificial emergency coping, acquiring an attribute element-screening value library, coping prompt information and coping behavior specification discrimination model corresponding to the coping strategy;

Acquiring personnel attribute information of a fourth trip personnel except the third trip personnel in the second trip personnel;

extracting attribute elements from the attribute information to obtain a plurality of attribute elements;

determining a screening value corresponding to the attribute element based on the attribute element-screening value library, and associating the screening value with the fourth trip personnel;

accumulating and calculating the screening value associated with the fourth trip personnel to obtain a screening value sum;

taking the largest screening value and the corresponding fourth trip personnel as fifth trip personnel;

controlling the tour guide robot to display the corresponding prompt information to the fifth trip personnel, and recording the current moment;

determining the second state behavior generated by the fifth traveler after the current moment and taking the second state behavior as a third state behavior;

based on the coping behavior specification judging model, coping behavior specification judgment is carried out on the third state behavior, and at least one non-specification item is obtained;

acquiring correction prompt information corresponding to the nonstandard item, and controlling the tour guide robot to display the correction prompt information to the fifth trip personnel;

and when the corresponding type corresponding to the corresponding strategy is non-manual emergency response, executing the corresponding strategy.

The invention provides an intelligent travel decision system based on big data, which comprises:

the acquisition module is used for acquiring a travel plan which is input by a user and needs to be subjected to travel decision;

the planning module is used for acquiring the travel big data and planning a travel guide suitable for the travel plan based on the travel big data;

and the pushing module is used for pushing the travel guide to a user as a travel decision result.

Preferably, the acquisition module performs the following operations:

obtaining a preset travel big data source set, wherein the travel big data source set comprises: a plurality of first sources;

obtaining a credit value corresponding to the first source;

if the credit value is greater than or equal to a preset credit threshold, the corresponding first source is used as a second source;

obtaining a source type corresponding to the second source, wherein the source type comprises: individuals and platforms;

when the source type corresponding to the second source is personal, acquiring a first data attribute of at least one first data item provided by the second source, and simultaneously acquiring information to be verified, provided by the second source, corresponding to the first data item;

acquiring at least one first verification policy corresponding to the first data attribute, and verifying the corresponding information to be verified based on the first verification policy;

When the first data item passes the verification, the corresponding first data item is taken as a second data item;

when the source type corresponding to the second source is a platform, acquiring a second data attribute of at least one third data item provided by the second source, and simultaneously acquiring a guarantee value corresponding to the second source for guaranteeing the third data item;

acquiring a guarantee threshold corresponding to the second data attribute, and taking the corresponding third data item as a fourth data item if the guarantee value is greater than or equal to the guarantee threshold;

and integrating the second data item and the fourth data item to obtain travel big data, and completing the acquisition.

Preferably, the acquisition module performs the following operations:

acquiring a plurality of credit records corresponding to the first source;

extracting the record characteristics of the credit records to obtain a plurality of record characteristics;

constructing a record characteristic-dynamic malicious value library, and determining a first dynamic malicious value corresponding to the record characteristic based on the record characteristic-dynamic malicious value library;

accumulating and calculating the first dynamic malicious value to obtain a malicious value sum;

acquiring the malicious value and a corresponding down-regulation strategy;

and acquiring a credit initial value corresponding to the first source, and performing down-regulation on the credit initial value based on the down-regulation strategy to acquire the credit value corresponding to the first source so as to finish the acquisition.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.

The technical scheme of the invention is further described in detail through the drawings and the embodiments.

Drawings

The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:

FIG. 1 is a flow chart of a big data based intelligent travel decision method in an embodiment of the invention;

FIG. 2 is a schematic diagram of an intelligent travel decision system based on big data according to an embodiment of the present invention.

Detailed Description

The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.

The invention provides an intelligent travel decision method based on big data, as shown in figure 1, comprising the following steps:

Step 1: acquiring a travel plan which is input by a user and needs to be subjected to travel decision;

step 2: acquiring tourist big data, and making a tourist guide suitable for the tourist plan based on the tourist big data;

step 3: pushing the tour guide to a user as a tour decision result.

The working principle and the beneficial effects of the technical scheme are as follows:

the user inputs a travel plan (e.g., 1 old man 2 adult 2 child travel to Nanjing and Suzhou travel 7 days) based on the intelligent terminal (e.g., smart phone and tablet, etc.); acquiring tourist big data (such as a strategy, an evaluation and a suitable crowd of tourist attractions in each place), and making a tourist guide suitable for a tourist plan based on the tourist big data (such as the first day, weather is fine, 1 adult accompanies the child to go to the amusement park with the highest score, 1 adult accompanies the old to go to the ancient urban area, and the like); the travel guide is pushed to the user.

The embodiment of the invention makes a proper travel guide according to the travel plan of the user based on the travel big data, helps the user to make a reasonable travel decision, improves convenience, and simultaneously improves the travel experience of the user.

The invention provides a big data-based intelligent travel decision method, wherein in the step 2, travel big data is acquired, and the method comprises the following steps:

Obtaining a preset travel big data source set, wherein the travel big data source set comprises: a plurality of first sources;

obtaining a credit value corresponding to the first source;

if the credit value is greater than or equal to a preset credit threshold, the corresponding first source is used as a second source;

obtaining a source type corresponding to the second source, wherein the source type comprises: individuals and platforms;

when the source type corresponding to the second source is personal, acquiring a first data attribute of at least one first data item provided by the second source, and simultaneously acquiring information to be verified, provided by the second source, corresponding to the first data item;

acquiring at least one first verification policy corresponding to the first data attribute, and verifying the corresponding information to be verified based on the first verification policy;

when the first data item passes the verification, the corresponding first data item is taken as a second data item;

when the source type corresponding to the second source is a platform, acquiring a second data attribute of at least one third data item provided by the second source, and simultaneously acquiring a guarantee value corresponding to the second source for guaranteeing the third data item;

Acquiring a guarantee threshold corresponding to the second data attribute, and taking the corresponding third data item as a fourth data item if the guarantee value is greater than or equal to the guarantee threshold;

and integrating the second data item and the fourth data item to obtain travel big data, and completing the acquisition.

The working principle and the beneficial effects of the technical scheme are as follows:

the first source corresponds to a travel platform, a travel attack website, etc.; preparing to obtain data provided by a first source (e.g., scenic spot aggression, scenic spot assessment, etc.); however, the authenticity of the data provided by the first source cannot be guaranteed, so that the credit verification is performed on the first source to obtain a credit value, the larger the credit value is, the larger the authenticity of the data provided by the first source historically is, and if the credit value is smaller than or equal to a preset credit threshold value, the corresponding first source is removed; secondly, the authenticity of the data to be provided after the rest of the second sources are eliminated can be verified; the source types of the second source are divided into individuals (e.g., tourists) and platforms (e.g., tourist platforms, etc.); when the source type is personal, acquiring first data attributes (such as scenic spot evaluation) of at least one first data line provided by the second source, acquiring information to be verified (such as scenic spot tickets) corresponding to first data items provided by the second source, acquiring first verification strategies (such as verifying whether to purchase scenic spot tickets for scenic spot evaluation) corresponding to the first data attributes, verifying the information to be verified based on the first verification strategies, and taking the corresponding first data items as second data items when verification is performed based on each first verification strategy; when the source type is the platform, the platform can verify the data such as evaluation (for example, the platform can evaluate the scenic spot only when buying a certain scenic spot ticket), so as to obtain the second data attribute (for example, scenic spot evaluation) of at least one third data item provided by the second source, and at the same time, obtain the guarantee value of the second source for guaranteeing the third data item (the larger the guarantee value, the larger the penalty if the provided data is not true); acquiring a guarantee threshold value corresponding to the second data attribute (the higher the importance of the data attribute is, the higher the requirement on the guarantee degree is, and if the guarantee value is larger than or equal to the guarantee threshold value, taking the corresponding third data item as a fourth data item; integrating the second data item and the third data item to obtain travel big data;

According to the embodiment of the invention, firstly, the credit of the first source is verified, secondly, the authenticity of the data provided by the second source is verified based on the difference of source types of the second source passing through the credit verification, and the travel big data is obtained by integrating the data passing through the authenticity verification, so that the reliability of the acquisition of the travel big data is ensured, and the method has more applicability under the current situation that the acquired data quality of the big data is different.

The invention provides a big data-based intelligent travel decision method, which comprises the steps of:

acquiring a plurality of credit records corresponding to the first source;

extracting the record characteristics of the credit records to obtain a plurality of record characteristics;

constructing a record characteristic-dynamic malicious value library, and determining a first dynamic malicious value corresponding to the record characteristic based on the record characteristic-dynamic malicious value library;

accumulating and calculating the first dynamic malicious value to obtain a malicious value sum;

acquiring the malicious value and a corresponding down-regulation strategy;

and acquiring a credit initial value corresponding to the first source, and performing down-regulation on the credit initial value based on the down-regulation strategy to acquire the credit value corresponding to the first source so as to finish the acquisition.

The working principle and the beneficial effects of the technical scheme are as follows:

when the corresponding credit value of the first source is acquired, comprehensive judgment is required to be carried out on the credit condition of the data provided by the first source in history; thus, a plurality of credit records corresponding to the first source are obtained, record characteristics are extracted, a first dynamic malicious value corresponding to the record characteristics is determined based on a constructed record characteristic-dynamic malicious value library, the larger the first dynamic malicious value is, the larger the corresponding record characteristics belong to bad credit characteristics (such as providing a non-scenic spot evaluation and the like), the first dynamic malicious value is calculated in an accumulated mode, a malicious value sum is obtained, a malicious value and a corresponding downregulation strategy (such as multiplying by 0.6, the larger the malicious value sum is, the larger the overall malicious degree is, the downregulation amplitude of the corresponding downregulation strategy is, the larger the credit initial value (such as 100) corresponding to the first source is obtained, the credit initial value is subjected to downregulation (such as 100 x 0.6=60) based on the downregulation strategy, and the credit value is obtained.

The invention provides a big data-based intelligent travel decision method, which constructs a record characteristic-dynamic malicious value library, and comprises the following steps:

acquiring a preset bad credit record characteristic set, wherein the record characteristic set comprises: a plurality of second recorded features;

Acquiring a result event corresponding to the second record feature, and simultaneously acquiring a feature weight of the second record feature corresponding to the result event;

based on a preset result severity judging model, carrying out result severity judgment on the result event to obtain a first severity value, and endowing the first severity value with the characteristic weight to obtain a to-be-counted value;

accumulating and calculating the to-be-counted value to obtain a to-be-paired value;

combining and pairing the value to be paired with the corresponding second record characteristic to obtain a pairing item;

acquiring a preset blank library, and storing the pairing items into the blank library;

and after the pairing items which are required to be stored in the blank library are stored, the blank library is used as a record characteristic-dynamic malicious value library, and the construction is completed.

The working principle and the beneficial effects of the technical scheme are as follows:

the malicious degree of different bad credit features needs to be comprehensively judged according to the results generated by the corresponding bad credit, but the generated results are updated continuously (for example, the evaluation of the bad scene points causes more and more tourists to step on the mine), so that dynamic judgment is needed to ensure the accuracy of the malicious degree judgment of the bad credit features; the second record feature corresponds to a bad credit feature (for example, providing an evaluation of a non-real scenic spot), obtaining a result event corresponding to the second record feature (for example, leading a tourist to step on a mine), obtaining a feature weight corresponding to the result event of the second record feature (for example, leading the tourist to step on the mine, if the reason leading the tourist to step on the mine is still other because of the provided evaluation of the non-real scenic spot, the feature weight is 1, if the reason leading the tourist to step on the mine is still other, the feature weight is less than 1), judging the severity of the result event based on a preset result severity judging model (a pre-trained model for judging the severity of the result event), obtaining a first severity value, giving the feature weight of the first severity value (multiplication of the two), and obtaining a waiting value; accumulating and counting the waiting count value to obtain a waiting pairing value, and carrying out combination pairing with the corresponding second record characteristic to obtain a pairing item; constructing a record characteristic-dynamic malicious value library based on the pairing items; the blank database is a blank database, and no content exists in the blank database.

The invention provides a big data-based intelligent travel decision method, in the step 2, a travel guide suitable for the travel plan is formulated based on the travel big data, and the method comprises the following steps:

training a travel guide to make a model;

and (3) setting a model based on the travel guide, and setting a travel guide suitable for the travel plan according to the travel big data.

The working principle and the beneficial effects of the technical scheme are as follows:

when the tour guide is formulated, a tour guide formulation model can be trained, the tour guide formulation model is based on the training tour guide formulation model, and the tour guide suitable for the tour plan is formulated according to the tour big data (for example, the travel of the old is represented in the tour plan, and the scenic spot, the play duration, the requirement for the middle-aged to accompany or not and the like which are most suitable for the old in the going-to place are determined based on the tour big data).

The invention provides a big data-based intelligent travel decision-making method, which is used for training a travel guide making model and comprises the following steps:

acquiring a plurality of first travel guide making events;

acquiring an experience value of a formulator corresponding to the first travel guide formulation event, and taking the corresponding first travel guide formulation event as a second travel guide formulation event if the experience value is greater than or equal to a preset experience threshold;

Acquiring a formulation process corresponding to the second travel guide formulation event;

splitting the formulation process into a plurality of first sub-processes;

sequencing the first sub-process according to a preset process sequencing rule to obtain a first process sequence;

extracting process characteristics of the first sub-process to obtain a plurality of first process characteristics;

acquiring a preset verification trigger feature library, matching the first process features with second process features in the verification trigger feature library, taking the corresponding first subprocess as a second subprocess if the matching is met, and simultaneously acquiring at least one second verification strategy corresponding to the matched second process features;

verifying the second sub-process based on the second verification policy, and taking the corresponding second verification policy as a third verification policy if the verification is not passed;

at least one influence item corresponding to the third verification policy is obtained, and meanwhile, an influence type corresponding to the influence item is obtained, wherein the influence type comprises: internal and external influences;

when the influence type of the influence item is an internal influence, a first influence value corresponding to the influence item is obtained, and the first influence value is marked on the second subprocess;

When the influence type of the influence item is external influence, acquiring a second influence value, an influence direction, an influence range and an influence target corresponding to the influence item;

determining the first sub-process corresponding to the influence target in the influence range in the influence direction of the second sub-process in the first process sequence, and taking the first sub-process as a third sub-process;

marking the second influence value on the third sub-process;

when the first influence value and the second influence value are marked, the first process sequence is used as a second process sequence;

accumulating the marked first influence value and the marked second influence value of each fourth sub-process in the second process sequence to obtain a sum of influence values;

acquiring the influence value and a corresponding process weight threshold value, and simultaneously acquiring a process weight of the fourth sub-process corresponding to the formulation process;

if the process weight is greater than or equal to the corresponding process weight threshold, eliminating the corresponding second travel guide making event;

when all the second travel guide making events needing to be removed are removed, taking the removed remaining second travel guide making events as third travel guide making events;

And performing model training on the third travel guide making event based on a preset model training algorithm to obtain a travel guide making model.

The working principle and the beneficial effects of the technical scheme are as follows:

when the model is established by training the tour guide, the tour guide suitable for the tour plan can be established based on the machine learning technology and how to learn the manual based on the tour big data; acquiring a plurality of first travel guide making events, wherein the first travel guide making events correspond to records of travel guides which are manually made suitable for travel planning based on travel big data; however, the manual making record is not necessarily reasonable, in order to ensure the training quality of the travel guide making model and improve the rationality of making the travel guide, the reliability verification needs to be performed on the first travel guide making event, and when the reliability verification is performed, the verification can be performed from the experience degree of the making party making the first travel guide and two dimensions of the specific making process, so that the experience value of the making party (at least one manual making person) corresponding to the first travel guide making event is obtained, the greater the experience value, the higher the experience richness of the making party is indicated, and if the experience value is greater than or equal to a preset experience threshold, the corresponding first travel guide making event is used as a second travel guide making event, and the specific making process verification is prepared; acquiring a formulation process corresponding to a second travel guide formulation event, splitting the formulation process into a plurality of first subprocesses, and sequencing the first subprocesses into a first process sequence according to a preset process sequencing rule (sequencing according to the sequence of the processes); extracting first process characteristics of a first sub-process, matching the first process characteristics with second process characteristics (such as how to determine suitable travel points of the old people) in a preset verification trigger characteristic library (a database storing process characteristics representing the sub-process to be verified), if the matching is met, indicating that the corresponding second sub-process needs to be further verified, verifying each first sub-process is not needed, reducing verification resources, meanwhile, acquiring a second verification strategy (such as how to determine whether the process of the suitable travel points of the old people is reasonable) corresponding to the matched second process characteristics, verifying the second sub-process, if the verification is not met, acquiring at least one influence item corresponding to a third verification strategy, wherein the influence types of the influence items are divided into internal influence (influence is caused on the current second sub-process, such as influence on rationality of the second sub-process) and external influence (influence is caused on other first sub-processes, such as unreasonable travel points of the old people of the second sub-process, stroke experience after the old people is influenced), and acquiring a physical strength corresponding value on the first influence when the influence types are internal influence; when the influence type is external influence, a corresponding second influence value, an influence direction (for example: back), an influence range (for example: 24 hours) and an influence target (for example: other strokes of the old) are obtained, a third sub-process is rapidly determined based on the influence direction, the influence range and the influence target, the external influence determination efficiency is improved, and the second influence value is marked on the third sub-process; accumulating and calculating a marked first influence value and a marked second influence value of each fourth sub-process in the second process sequence to obtain a sum of the influence values; acquiring a process weight of the fourth sub-process corresponding to the formulation process, wherein the process weight is larger, the process importance is larger (in general, the more the formulation process of the travel guide is forward, the more the effect on the following formulation process is exerted, and therefore, the more the process is forward, the process weight is larger); acquiring an influence value and a corresponding process weight threshold value, wherein the larger the influence value is, the higher the process non-standardization degree is, the larger the requirement on the process weight is, the smaller the process weight threshold value is, and if the process weight is greater than or equal to the preset process weight threshold value, the corresponding second travel guide making event is eliminated; based on a preset model training algorithm (for example, a machine learning algorithm), performing model training on the third travel guide making event with the remainder removed (the model training belongs to the category of the prior art and is not described in detail), so as to obtain a travel guide making model;

The influence value and the corresponding process weight threshold value are obtained through the following formula:

wherein,and for the influence value and the corresponding process weight threshold value, mu is a preset relation coefficient, and alpha is the influence value sum.

The invention provides an intelligent travel decision method based on big data, which further comprises the following steps:

extracting a travel route from the travel guide, wherein the travel route passes through at least one first scenic spot;

determining at least one first emergency corresponding to the first scenic spot from the tourist big data;

based on a preset severity judging model, carrying out severity judgment on the first sudden event, obtaining a second severity value, and associating with the corresponding first scenic spot;

accumulating and calculating the second serious value associated with the first scenic spot to obtain a serious value sum;

if the serious value sum is greater than or equal to a preset serious value sum and a threshold value sum, taking the corresponding first scenic spot as a second scenic spot, and taking the first emergency corresponding to the second scenic spot as a second emergency;

acquiring at least one first state behavior corresponding to the second emergency;

extracting at least one first travel person corresponding to the second scenic spot from the travel guide;

When the first traveler enters the second Jing Oushi, taking the first traveler entering the second scenic spot as a second traveler;

controlling a tour guide trolley in the second scenic spot to follow a tour team consisting of the second tour personnel;

in the following process, controlling the tour guide trolley to dynamically acquire a plurality of second state behaviors generated by the second trip personnel;

matching the second state behavior with the first state behavior, if the matching is met, acquiring the characterization degree corresponding to the first state behavior which is met by the matching, and correlating with the corresponding second emergency;

accumulating and calculating the characterization degree associated with the second emergency event to obtain a characterization degree sum;

acquiring a characterization degree and a threshold value corresponding to the second emergency, and taking the corresponding second emergency as a third emergency and taking the corresponding second trip personnel as a third trip personnel if the characterization degree and the threshold value are larger than or equal to each other;

at least one coping strategy corresponding to the third emergency is obtained, and meanwhile, a coping type corresponding to the coping strategy is obtained, wherein the coping type comprises: manual emergency handling and non-manual emergency handling;

When the coping type corresponding to the coping strategy is artificial emergency coping, acquiring an attribute element-screening value library, coping prompt information and coping behavior specification discrimination model corresponding to the coping strategy;

acquiring personnel attribute information of a fourth trip personnel except the third trip personnel in the second trip personnel;

extracting attribute elements from the attribute information to obtain a plurality of attribute elements;

determining a screening value corresponding to the attribute element based on the attribute element-screening value library, and associating the screening value with the fourth trip personnel;

accumulating and calculating the screening value associated with the fourth trip personnel to obtain a screening value sum;

taking the largest screening value and the corresponding fourth trip personnel as fifth trip personnel;

controlling the tour guide robot to display the corresponding prompt information to the fifth trip personnel, and recording the current moment;

determining the second state behavior generated by the fifth traveler after the current moment and taking the second state behavior as a third state behavior;

based on the coping behavior specification judging model, coping behavior specification judgment is carried out on the third state behavior, and at least one non-specification item is obtained;

Acquiring correction prompt information corresponding to the nonstandard item, and controlling the tour guide robot to display the correction prompt information to the fifth trip personnel;

and when the corresponding type corresponding to the corresponding strategy is non-manual emergency response, executing the corresponding strategy.

The working principle and the beneficial effects of the technical scheme are as follows:

when tourists enter special scenic spots (such as a highland scenic spot, a recreation ground and the like), adverse reactions (such as altitude reactions and discomfort after sitting on a roller coaster) can occur, because the special scenic spots generally occupy larger area, if the adverse reactions are serious, rescue workers can not arrive at the scene in time to rescue, and simple rescue is required by the peers, but the peers can not know how to rescue, so that the adverse reactions can be more serious;

therefore, firstly, determining that a user enters the special scenic spot, extracting a travel route from a travel guide, wherein the travel route passes through the first scenic spot, determining a first emergency corresponding to the first scenic spot from travel big data (for example, tourists generate altitude reaction and the like), and judging the severity of the first emergency based on a preset severity judging model (a pre-trained model for judging the severity of the emergency), so as to obtain a second severity value; accumulating and calculating a second serious value associated with the first scenic spot to obtain a serious value sum, if the serious value sum is greater than or equal to a preset serious value and a threshold value, indicating that a user possibly generates adverse reaction in the corresponding first scenic spot, and determining a second emergency corresponding to the second scenic spot as the second scenic spot, wherein the second emergency is used as adverse reaction pre-judgment, and acquiring a first state behavior (adverse reaction behavior) corresponding to the second emergency; extracting a first traveler who goes to a second scenic spot from the travel guide, but specifically identifying the second traveler who goes to the second scenic spot; the second trip staff arrives at the entrance of the second scenic spot, and the tour guide trolley follows the trip team formed by the second trip staff;

Secondly, whether adverse reactions are generated by users or not needs to be confirmed; controlling the tour guide trolley to dynamically acquire a plurality of second state behaviors (actions, expressions, languages and the like) generated by a second person, matching the second state behaviors with the first state behaviors by setting a camera, and if the second state behaviors match, acquiring the characterization degree corresponding to the matched first state behaviors, wherein the characterization degree is larger, and indicating that the generated state behaviors generate adverse reactions on the surface of a user; accumulating and calculating the associated characterization degree of the second emergency, and obtaining a characterization degree sum, wherein if the characterization degree sum is greater than or equal to the characterization degree and the threshold value corresponding to the second emergency, the corresponding third emergency is indicated to occur in the corresponding third trip personnel;

then, the user who generates adverse reaction needs to be rescued; at least one coping strategy corresponding to the third emergency is obtained (for example, a companion brings a third trip person to a quiet place to lie down, eyes are closed and tranquilize, the head is deviated to one side so as to prevent vomit from choking into the air pipe to cause choking, and meanwhile, deep breathing is carried out), and coping types of the coping strategy are divided into manual emergency coping (for example, companion assistance) and non-manual emergency coping (for example, a latest rescue site is notified); however, when the coping types corresponding to the coping strategies are artificial urgent coping, the number of the companions is plural, and the companions possibly discuss who is helped by them, so that time is wasted and the rescue efficiency is reduced, therefore, a property element-screening value library (comprising screening values corresponding to property elements of a proper rescue person, the larger the screening value is, the more proper) corresponding to the coping strategy is obtained, and the coping prompt information (for example, a model for carrying a third trip person to lie down in a quiet place, keeping eyes in mind, biasing the head to one side so as to prevent vomit from choking the air pipe to cause choking, and meanwhile, deeply breathing) and a behavior specification judging model (a model for judging whether the rescue behavior generated by the rescue companions is normal or not trained in advance) are obtained; acquiring personnel attribute information of a fourth trip personnel except the third trip personnel, extracting attribute elements (such as physical strength and the like), and determining a screening value based on an attribute element-screening value library; accumulating and calculating screening values associated with fourth travelers to obtain screening value sums, wherein the larger the screening value sums are, the more suitable the corresponding fifth travelers are, controlling the tour guide robot to display response prompt information to the fifth travelers, then the fifth travelers should go to rescue the third travelers, determining third state behaviors generated at the current moment, judging the third state behaviors according to the response behavior standard on the basis of a response behavior standard judging model, obtaining at least one non-standard item (for example, not lying down), obtaining correction prompt information (for example, enable the opposite side to lie down) corresponding to the non-standard item, displaying the correction prompt information to the fifth travelers, and reminding the fifth travelers to correct the correction; when the coping type is non-manual emergency coping, the coping type is directly executed;

Considering that adverse reactions can be generated when tourists play in special scenic spots, the embodiment of the invention firstly confirms whether the user can pass, monitors the state of the tourists by the docking tour guide trolley when the user passes, screens out the best companion when the adverse reactions are generated, and helps the user who generates the adverse reactions, thereby prompting the help efficiency and being more intelligent.

The invention provides an intelligent travel decision system based on big data, as shown in fig. 2, comprising:

the acquisition module 1 is used for acquiring a travel plan which is input by a user and needs to be subjected to travel decision;

a formulation module 2, configured to obtain travel big data, and based on the travel big data, formulate a travel guide suitable for the travel plan;

and the pushing module 3 is used for pushing the travel guide to the user as a travel decision result.

The working principle and the beneficial effects of the technical scheme are described in the method claims and are not repeated.

It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.

Claims (10)

1. An intelligent travel decision method based on big data is characterized by comprising the following steps:

step 1: acquiring a travel plan which is input by a user and needs to be subjected to travel decision;

step 2: acquiring tourist big data, and making a tourist guide suitable for the tourist plan based on the tourist big data;

step 3: pushing the tour guide to a user as a tour decision result.

2. The intelligent travel decision-making method based on big data as recited in claim 1, wherein in said step 2, obtaining travel big data comprises:

obtaining a preset travel big data source set, wherein the travel big data source set comprises: a plurality of first sources;

obtaining a credit value corresponding to the first source;

if the credit value is greater than or equal to a preset credit threshold, the corresponding first source is used as a second source;

obtaining a source type corresponding to the second source, wherein the source type comprises: individuals and platforms;

when the source type corresponding to the second source is personal, acquiring a first data attribute of at least one first data item provided by the second source, and simultaneously acquiring information to be verified, provided by the second source, corresponding to the first data item;

Acquiring at least one first verification policy corresponding to the first data attribute, and verifying the corresponding information to be verified based on the first verification policy;

when the first data item passes the verification, the corresponding first data item is taken as a second data item;

when the source type corresponding to the second source is a platform, acquiring a second data attribute of at least one third data item provided by the second source, and simultaneously acquiring a guarantee value corresponding to the second source for guaranteeing the third data item;

acquiring a guarantee threshold corresponding to the second data attribute, and taking the corresponding third data item as a fourth data item if the guarantee value is greater than or equal to the guarantee threshold;

and integrating the second data item and the fourth data item to obtain travel big data, and completing the acquisition.

3. The intelligent travel decision method based on big data as recited in claim 2, wherein obtaining the credit value corresponding to the first source comprises:

acquiring a plurality of credit records corresponding to the first source;

extracting the record characteristics of the credit records to obtain a plurality of record characteristics;

constructing a record characteristic-dynamic malicious value library, and determining a first dynamic malicious value corresponding to the record characteristic based on the record characteristic-dynamic malicious value library;

Accumulating and calculating the first dynamic malicious value to obtain a malicious value sum;

acquiring the malicious value and a corresponding down-regulation strategy;

and acquiring a credit initial value corresponding to the first source, and performing down-regulation on the credit initial value based on the down-regulation strategy to acquire the credit value corresponding to the first source so as to finish the acquisition.

4. A method of intelligent travel decision based on big data as recited in claim 3, wherein constructing a record feature-dynamic malicious value library comprises:

acquiring a preset bad credit record characteristic set, wherein the record characteristic set comprises: a plurality of second recorded features;

acquiring a result event corresponding to the second record feature, and simultaneously acquiring a feature weight of the second record feature corresponding to the result event;

based on a preset result severity judging model, carrying out result severity judgment on the result event to obtain a first severity value, and endowing the first severity value with the characteristic weight to obtain a to-be-counted value;

accumulating and calculating the to-be-counted value to obtain a to-be-paired value;

combining and pairing the value to be paired with the corresponding second record characteristic to obtain a pairing item;

Acquiring a preset blank library, and storing the pairing items into the blank library;

and after the pairing items which are required to be stored in the blank library are stored, the blank library is used as a record characteristic-dynamic malicious value library, and the construction is completed.

5. The intelligent travel decision-making method based on big data as recited in claim 1, wherein in said step 2, a travel guide suitable for said travel plan is formulated based on said big travel data, comprising:

training a travel guide to make a model;

and (3) setting a model based on the travel guide, and setting a travel guide suitable for the travel plan according to the travel big data.

6. The big data based intelligent travel decision method of claim 5, wherein training the travel guideline formulation model comprises:

acquiring a plurality of first travel guide making events;

acquiring an experience value of a formulator corresponding to the first travel guide formulation event, and taking the corresponding first travel guide formulation event as a second travel guide formulation event if the experience value is greater than or equal to a preset experience threshold;

acquiring a formulation process corresponding to the second travel guide formulation event;

Splitting the formulation process into a plurality of first sub-processes;

sequencing the first sub-process according to a preset process sequencing rule to obtain a first process sequence;

extracting process characteristics of the first sub-process to obtain a plurality of first process characteristics;

acquiring a preset verification trigger feature library, matching the first process features with second process features in the verification trigger feature library, taking the corresponding first subprocess as a second subprocess if the matching is met, and simultaneously acquiring at least one second verification strategy corresponding to the matched second process features;

verifying the second sub-process based on the second verification policy, and taking the corresponding second verification policy as a third verification policy if the verification is not passed;

at least one influence item corresponding to the third verification policy is obtained, and meanwhile, an influence type corresponding to the influence item is obtained, wherein the influence type comprises: internal and external influences;

when the influence type of the influence item is an internal influence, a first influence value corresponding to the influence item is obtained, and the first influence value is marked on the second subprocess;

When the influence type of the influence item is external influence, acquiring a second influence value, an influence direction, an influence range and an influence target corresponding to the influence item;

determining the first sub-process corresponding to the influence target in the influence range in the influence direction of the second sub-process in the first process sequence, and taking the first sub-process as a third sub-process;

marking the second influence value on the third sub-process;

when the first influence value and the second influence value are marked, the first process sequence is used as a second process sequence;

accumulating the marked first influence value and the marked second influence value of each fourth sub-process in the second process sequence to obtain a sum of influence values;

acquiring the influence value and a corresponding process weight threshold value, and simultaneously acquiring a process weight of the fourth sub-process corresponding to the formulation process;

if the process weight is greater than or equal to the corresponding process weight threshold, eliminating the corresponding second travel guide making event;

when all the second travel guide making events needing to be removed are removed, taking the removed remaining second travel guide making events as third travel guide making events;

And performing model training on the third travel guide making event based on a preset model training algorithm to obtain a travel guide making model.

7. The intelligent travel decision-making method based on big data as recited in claim 1, further comprising:

extracting a travel route from the travel guide, wherein the travel route passes through at least one first scenic spot;

determining at least one first emergency corresponding to the first scenic spot from the tourist big data;

based on a preset severity judging model, carrying out severity judgment on the first sudden event, obtaining a second severity value, and associating with the corresponding first scenic spot;

accumulating and calculating the second serious value associated with the first scenic spot to obtain a serious value sum;

if the serious value sum is greater than or equal to a preset serious value sum and a threshold value sum, taking the corresponding first scenic spot as a second scenic spot, and taking the first emergency corresponding to the second scenic spot as a second emergency;

acquiring at least one first state behavior corresponding to the second emergency;

extracting at least one first travel person corresponding to the second scenic spot from the travel guide;

When the first traveler enters the second Jing Oushi, taking the first traveler entering the second scenic spot as a second traveler;

controlling a tour guide trolley in the second scenic spot to follow a tour team consisting of the second tour personnel;

in the following process, controlling the tour guide trolley to dynamically acquire a plurality of second state behaviors generated by the second trip personnel;

matching the second state behavior with the first state behavior, if the matching is met, acquiring the characterization degree corresponding to the first state behavior which is met by the matching, and correlating with the corresponding second emergency;

accumulating and calculating the characterization degree associated with the second emergency event to obtain a characterization degree sum;

acquiring a characterization degree and a threshold value corresponding to the second emergency, and taking the corresponding second emergency as a third emergency and taking the corresponding second trip personnel as a third trip personnel if the characterization degree and the threshold value are larger than or equal to each other;

at least one coping strategy corresponding to the third emergency is obtained, and meanwhile, a coping type corresponding to the coping strategy is obtained, wherein the coping type comprises: manual emergency handling and non-manual emergency handling;

When the coping type corresponding to the coping strategy is artificial emergency coping, acquiring an attribute element-screening value library, coping prompt information and coping behavior specification discrimination model corresponding to the coping strategy;

acquiring personnel attribute information of a fourth trip personnel except the third trip personnel in the second trip personnel;

extracting attribute elements from the attribute information to obtain a plurality of attribute elements;

determining a screening value corresponding to the attribute element based on the attribute element-screening value library, and associating the screening value with the fourth trip personnel;

accumulating and calculating the screening value associated with the fourth trip personnel to obtain a screening value sum;

taking the largest screening value and the corresponding fourth trip personnel as fifth trip personnel;

controlling the tour guide robot to display the corresponding prompt information to the fifth trip personnel, and recording the current moment;

determining the second state behavior generated by the fifth traveler after the current moment and taking the second state behavior as a third state behavior;

based on the coping behavior specification judging model, coping behavior specification judgment is carried out on the third state behavior, and at least one non-specification item is obtained;

Acquiring correction prompt information corresponding to the nonstandard item, and controlling the tour guide robot to display the correction prompt information to the fifth trip personnel;

and when the corresponding type corresponding to the corresponding strategy is non-manual emergency response, executing the corresponding strategy.

8. An intelligent travel decision system based on big data, comprising:

the acquisition module is used for acquiring a travel plan which is input by a user and needs to be subjected to travel decision;

the planning module is used for acquiring the travel big data and planning a travel guide suitable for the travel plan based on the travel big data;

and the pushing module is used for pushing the travel guide to a user as a travel decision result.

9. The intelligent travel decision system based on big data as recited in claim 8, wherein said acquisition module performs the following operations:

obtaining a preset travel big data source set, wherein the travel big data source set comprises: a plurality of first sources;

obtaining a credit value corresponding to the first source;

if the credit value is greater than or equal to a preset credit threshold, the corresponding first source is used as a second source;

obtaining a source type corresponding to the second source, wherein the source type comprises: individuals and platforms;

When the source type corresponding to the second source is personal, acquiring a first data attribute of at least one first data item provided by the second source, and simultaneously acquiring information to be verified, provided by the second source, corresponding to the first data item;

acquiring at least one first verification policy corresponding to the first data attribute, and verifying the corresponding information to be verified based on the first verification policy;

when the first data item passes the verification, the corresponding first data item is taken as a second data item;

when the source type corresponding to the second source is a platform, acquiring a second data attribute of at least one third data item provided by the second source, and simultaneously acquiring a guarantee value corresponding to the second source for guaranteeing the third data item;

acquiring a guarantee threshold corresponding to the second data attribute, and taking the corresponding third data item as a fourth data item if the guarantee value is greater than or equal to the guarantee threshold;

and integrating the second data item and the fourth data item to obtain travel big data, and completing the acquisition.

10. The big data based intelligent travel decision system of claim 9, wherein the acquisition module performs the following operations:

Acquiring a plurality of credit records corresponding to the first source;

extracting the record characteristics of the credit records to obtain a plurality of record characteristics;

constructing a record characteristic-dynamic malicious value library, and determining a first dynamic malicious value corresponding to the record characteristic based on the record characteristic-dynamic malicious value library;

accumulating and calculating the first dynamic malicious value to obtain a malicious value sum;

acquiring the malicious value and a corresponding down-regulation strategy;

and acquiring a credit initial value corresponding to the first source, and performing down-regulation on the credit initial value based on the down-regulation strategy to acquire the credit value corresponding to the first source so as to finish the acquisition.