CN117540098A - Travel recommendation method and device, electronic equipment and storage medium - Google Patents
Travel recommendation method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117540098A CN117540098A CN202311618297.3A CN202311618297A CN117540098A CN 117540098 A CN117540098 A CN 117540098A CN 202311618297 A CN202311618297 A CN 202311618297A CN 117540098 A CN117540098 A CN 117540098A
- Authority
- CN
- China
- Prior art keywords
- travel
- data
- recommendation
- sample
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000012549 training Methods 0.000 claims abstract description 33
- 230000006399 behavior Effects 0.000 claims description 29
- 230000006870 function Effects 0.000 claims description 29
- 230000003993 interaction Effects 0.000 claims description 25
- 238000004590 computer program Methods 0.000 claims description 13
- 230000002452 interceptive effect Effects 0.000 description 10
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000013527 convolutional neural network Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000011478 gradient descent method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 230000002787 reinforcement Effects 0.000 description 2
- 210000004556 brain Anatomy 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 235000006694 eating habits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/14—Travel agencies
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a travel recommendation method, a travel recommendation device, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring real-time data of tourist attractions and destination recommendation data; inputting the real-time data of the tourist attractions and the destination recommendation data into a tourist recommendation model, and obtaining and outputting tourist suggestions by the tourist recommendation model; the travel recommendation model is obtained based on sample real-time data of tourist attractions, sample destination recommendation data and label travel advice training. On one hand, according to different travel destinations, destination recommendation data are different, and then travel recommendation is performed based on the destination recommendation data, so that individuation and accuracy of travel recommendation are improved; on the other hand, the real-time data of tourist attractions is referenced, so that the experience of travelers is optimized, more accurate travel advice is provided, the travelers are helped to plan the journey better, queuing and congestion are avoided, and the specific requirements and preferences of the travelers are fully met.
Description
Technical Field
The invention relates to the technical field of travel recommendation, in particular to a travel recommendation method, a travel recommendation device, electronic equipment and a storage medium.
Background
With the continuous improvement of information technology and the continuous development of the tourism industry, the information technology brings new opportunities for the development of the tourism industry. Strengthen informatization construction and be favorable to promoting the competitiveness of tourism platform, the tourism platform that informatization level is high moreover can provide more comprehensive sight spot information for the visitor, provides better tourism suggestion for the visitor to improve the experience sense of visitor.
In the prior art, the traditional travel strategies generally lack individuation, cannot adapt to the characteristics and the preferences of different travelers, and cannot consider real-time changes, so that the experience of the travelers is poor and the efficiency is low.
Disclosure of Invention
The invention provides a travel recommendation method, a device, electronic equipment and a storage medium, which are used for solving the defects that the traditional travel strategy in the prior art is generally lack of individuation, cannot adapt to the characteristics and the preference of different travelers, and cannot consider real-time change, so that the experience of the travelers is poor and the efficiency is low.
The invention provides a travel recommendation method, which comprises the following steps:
acquiring real-time data of tourist attractions and destination recommendation data;
inputting the real-time data of the tourist attractions and the destination recommendation data into a tourist recommendation model, and obtaining and outputting tourist suggestions by the tourist recommendation model;
the travel recommendation model is obtained based on sample real-time data of tourist attractions, sample destination recommendation data and label travel advice training.
According to the travel recommendation method provided by the invention, the acquisition steps of the travel recommendation model comprise:
acquiring an initial travel recommendation model, sample real-time data of the tourist attraction, sample destination recommendation data and the label travel suggestion;
inputting the sample real-time data and the sample destination recommendation data into the initial travel recommendation model, and obtaining and outputting predicted travel suggestions by the initial travel recommendation model;
and determining a loss function value based on the difference between the predicted travel advice and the tag travel advice, and carrying out parameter iteration on the initial travel recommendation model based on the loss function value to obtain the travel recommendation model.
According to the travel recommendation method provided by the invention, the training steps of the initial travel recommendation model comprise:
training a user model based on the user's sample historical behavior data, sample historical preference data, and user travel preference tags;
training the user model based on sample preference data, sample demand data and user travel demand labels of the user to obtain the initial travel recommendation model.
According to the travel recommendation method provided by the invention, the steps of acquiring the sample preference data and the sample demand data of the user comprise the following steps:
acquiring the sample historical behavior data and the sample historical preference data;
and extracting the sample preference data and the sample demand data from the sample history behavior data and the sample history preference data.
According to the travel recommendation method provided by the invention, the real-time data of the tourist attraction and the destination recommendation data are input into a travel recommendation model, travel suggestions are obtained and output by the travel recommendation model, and then the method further comprises the following steps:
acquiring feedback data of a user;
and carrying out parameter adjustment on the travel recommendation model based on the feedback data of the user.
According to the travel recommendation method provided by the invention, the parameter adjustment is performed on the travel recommendation model based on the feedback data of the user, and the method comprises the following steps:
determining the frequency and depth of the interaction topics in the feedback data of the user;
and carrying out parameter adjustment on the travel recommendation model based on the interaction topic frequency and the interaction topic depth.
The invention also provides a travel recommendation device, which comprises:
the acquisition unit is used for acquiring real-time data of tourist attractions and destination recommendation data;
the tourist recommendation unit is used for inputting the real-time data of the tourist attraction and the destination recommendation data into a tourist recommendation model, and obtaining and outputting tourist recommendation by the tourist recommendation model;
the travel recommendation model is obtained based on sample real-time data of tourist attractions, sample destination recommendation data and label travel advice training.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the travel recommendation method as described in any one of the above when executing the program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a travel recommendation method as described in any of the above.
The present invention also provides a computer program product comprising a computer program which when executed by a processor implements a travel recommendation method as described in any one of the above.
According to the travel recommendation method, the device, the electronic equipment and the storage medium, real-time data and destination recommendation data of a tourist attraction are obtained, the real-time data and the destination recommendation data of the tourist attraction are input into a travel recommendation model, travel suggestions are obtained and output by the travel recommendation model, and the travel recommendation model is obtained based on sample real-time data and sample destination recommendation data of the tourist attraction and label travel suggestion training. On one hand, according to different travel destinations, destination recommendation data are different, and then travel recommendation is performed based on the destination recommendation data, so that individuation and accuracy of travel recommendation are improved; on the other hand, the real-time data of tourist attractions is referenced, so that the experience of travelers is optimized, more accurate travel advice is provided, the travelers are helped to plan the journey better, queuing and congestion are avoided, and the specific requirements and preferences of the travelers are fully met.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a travel recommendation method provided by the invention;
FIG. 2 is a second flow chart of the travel recommendation method according to the present invention;
FIG. 3 is a schematic view of a travel recommendation device according to the present invention;
fig. 4 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The terms first, second and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate, such that embodiments of the present application may be capable of being practiced in sequences other than those illustrated and described herein, and that "first," "second," etc. are typically of the same type.
In the related art, a personalized recommendation system provides personalized suggestions for a user, including commodities, music, movies, and the like, according to the historical behaviors and preferences of the user. Some applications and platforms provide real-time data such as real-time traffic information, weather conditions, social media updates, and the like. These data are used to provide real-time information and services. Deep learning technology, deep learning is a machine learning technology that simulates the learning process of the human brain through a neural network to identify patterns and make predictions.
Some location-based recommendation systems use the user's location information and historical data to recommend nearby attractions, restaurants, etc. These systems can provide personalized advice, but often lack deep learning applications and comprehensive consideration of scenic spot heat, traffic conditions.
Based on the above-mentioned problems, the present invention provides a travel recommendation method, and fig. 1 is a schematic flow chart of the travel recommendation method provided by the present invention, as shown in fig. 1, the method includes:
step 110, acquiring real-time data of tourist attractions and destination recommendation data.
In particular, given that the main problem of traditional travel attack generation methods is the lack of personalization, the attacks they provide are generic, failing to fully take into account the characteristics and preferences of the traveler, resulting in the traveler not being able to obtain advice that exactly matches his needs; not adapt to real-time changes: conventional methods typically rely on static data and cannot accommodate changes in real-time conditions, meaning that they do not provide real-time important information for the day; lack of comprehensiveness: conventional attack-generation methods often fail to provide comprehensive information about sight queuing, congestion, and real-time traffic conditions, which makes travelers lack comprehensive and accurate information in planning trips.
Accordingly, real-time data of the tourist attraction and destination recommended data can be obtained, wherein the real-time data of the tourist attraction can comprise real-time traffic conditions, real-time attraction heat, real-time queuing time length and real-time congestion information of the tourist attraction.
Here, the real-time traffic condition may include public traffic condition, road congestion condition, etc., which is not particularly limited by the embodiment of the present invention. The real-time scenic spot heat is used for reflecting the real-time heat of scenic spots and is determined according to the number of tourists, evaluation and other data. The real-time queuing time length is used for reflecting the real-time queuing condition of the scenic spot. The real-time congestion information is used to reflect the congestion areas and conditions in the city.
Destination recommendation data, i.e., recommendation information related to travel destinations, may include cultures, delicacies, scenic spots, etc. of the destination, for example, histories, cultural heritage, famous scenic spots, special catering, etc., to which embodiments of the present invention are not limited in detail.
It can be understood that, according to different travel destinations, destination recommendation data are different, and the follow-up travel recommendation is performed based on the destination recommendation data, so that individuation and accuracy of the travel recommendation are improved.
Step 120, inputting the real-time data of the tourist attraction and the destination recommendation data into a tourist recommendation model, and obtaining and outputting a tourist recommendation by the tourist recommendation model;
the travel recommendation model is obtained based on sample real-time data of tourist attractions, sample destination recommendation data and label travel advice training.
Specifically, after the real-time data and the destination recommendation data of the tourist attraction are acquired, the real-time data and the destination recommendation data of the tourist attraction may be input into the tourist recommendation model, and the tourist recommendation model obtains and outputs the tourist advice.
Here, the travel advice may include scenic spot advice, dining advice, cultural experience advice, reference information about scenic spot heat, queuing time, and congestion, and the like, which is not particularly limited in the embodiment of the present invention.
Here, the travel recommendation model may be a multi-layer convolutional neural network (Convolutional Neural Network, CNN) with a cascade structure, a cyclic neural network (Recurrent Neural Network, RNN), a combination structure of a CNN model, an RNN model, an attention model and a deep reinforcement learning model, or the like, which is not particularly limited in the embodiment of the present invention.
For example, convolutional neural networks may be used to learn complex relationships between users and travel elements; the recurrent neural network can be used for modeling of historical behavior of the user; the attention model is used for capturing important elements focused in real-time data and destination recommendation data of tourist attractions and is used for personalized recommendation; the deep reinforcement learning model can learn complex relations between users and travel items, and accuracy of recommendation is improved.
In order to be able to better generate travel advice, a travel recommendation model needs to be trained by the following steps:
first, an initial travel recommendation model may be pre-built, and sample real-time data of tourist attractions, sample destination recommendation data, and tag travel advice may also be pre-collected.
Here, the initial travel recommendation model may be trained based on the user's sample preference data and sample demand data.
After the initial travel recommendation model is obtained, the sample real-time data, sample destination recommendation data and label travel recommendation data of the collected tourist attractions in advance can be applied to train the initial travel recommendation model:
firstly, sample real-time data and sample destination recommendation data of tourist attractions are input into an initial tourist recommendation model, and a predicted tourist recommendation is obtained and output by the initial tourist recommendation model.
After the predicted travel advice is obtained based on the initial travel advice model, the predicted travel advice and the pre-collected label travel advice can be compared, a loss function value is calculated according to the difference degree between the predicted travel advice and the pre-collected label travel advice, parameter iteration is carried out on the initial travel advice model based on the loss function value, and the initial travel advice model after parameter iteration is completed is recorded as the travel advice model.
It will be appreciated that the greater the degree of difference between the predicted travel advice and the pre-collected labeled travel advice, the greater the loss function value; the smaller the degree of difference between the predicted travel advice and the pre-collected labeled travel advice, the smaller the loss function value.
That is, in the training process of the initial travel recommendation model, the function of generating travel advice based on the real-time data of the tourist attractions and the destination recommendation data is learned.
Here, the cross entropy loss function (Cross Entropy Loss Function), the mean square error loss function (Mean Squared Error, MSE) and the random gradient descent method may be used to update the parameters of the initial travel recommendation model, which is not specifically limited in the embodiment of the present invention.
Further, after generating the travel advice based on the travel recommendation model, the generated travel advice may be communicated to a user interface where the user may view the generated travel advice in the application. The user interface may include more setup options to allow users to more finely configure their travel information and needs, thereby increasing more flexibility and personalization.
In addition, the generation of the travel advice can be more diversified, so that the user has more choices, and the satisfaction degree of the user is improved.
Accordingly, the user interface may include a recommendation diversity module to ensure that travel advice is generated not only for the user's interests, but also to include a number of diversity advice to allow the user more options.
The method provided by the embodiment of the invention acquires the real-time data of the tourist attractions and the destination recommendation data, inputs the real-time data of the tourist attractions and the destination recommendation data into a tourist recommendation model, and obtains and outputs tourist suggestions by the tourist recommendation model, wherein the tourist recommendation model is obtained based on sample real-time data of the tourist attractions, sample destination recommendation data and label tourist suggestion training. On one hand, according to different travel destinations, destination recommendation data are different, and then travel recommendation is performed based on the destination recommendation data, so that individuation and accuracy of travel recommendation are improved; on the other hand, the real-time data of tourist attractions is referenced, so that the experience of travelers is optimized, more accurate travel advice is provided, the travelers are helped to plan the journey better, queuing and congestion are avoided, and the specific requirements and preferences of the travelers are fully met.
The personalized travel advice generated by the rule-based travel advice algorithm in the prior art may be easier to interpret and adjust, but the travel advice generated in the complex situation may not meet the requirements and preferences of the user, and compared with the rule-based travel advice algorithm, the method provided by the embodiment of the invention can provide more accurate personalized travel advice in the complex situation, and meets the requirements and preferences of the user.
Based on the above embodiment, the training step of the initial travel recommendation model includes:
step 210, training a user model based on the sample historical behavior data, sample historical preference data, and the user travel preference tags of the user;
step 220, training the user model based on the sample preference data, sample demand data and the user travel demand label to obtain the initial travel recommendation model.
Specifically, the user model may be trained in advance, and sample historical behavior data, sample historical preference data, and user travel preference tags of the user may be collected in advance, prior to obtaining the initial travel recommendation model.
Here, the sample historical behavior data of the user may be a place where the user has visited, a selected traffic mode, a stay time, etc., and these information may provide insight about travel preference and habit of the user, or may be actions such as praise, collection, scoring, etc. of the user in the application program, which is not limited in particular by the embodiment of the present invention.
Here, the sample historical preference data of the user is used to reflect preferences such as travel behavior, tourist attractions, and dining records during the user's historical travel.
The initial user model may be pre-built, where parameters of the initial user model may be randomly generated or preset, and the embodiment of the present invention is not limited in particular.
The initial user model here may be a deep neural network (Deep Neural Networks, DNN), a combination structure of CNN and DNN, etc., which is not particularly limited in the embodiment of the present invention.
After the initial user model is obtained, the collected sample historical behavior data, sample historical preference data and user travel preference labels of the user can be applied to train the initial user model:
first, the sample historical behavior data and the sample historical preference data of the user can be input into an initial user model, and the user predicted travel preference can be obtained and output by the initial user model.
After obtaining the user predicted travel preference based on the initial user model, the user predicted travel preference can be compared with the user travel preference labels collected in advance, a first loss function value is calculated according to the difference degree between the user predicted travel preference and the user travel preference labels, parameter iteration is carried out on the initial user model based on the first loss function value, and the initial user model after parameter iteration is completed is recorded as a user model.
It will be appreciated that the greater the degree of difference between the user predicted travel preference and the pre-collected user travel preference tags, the greater the first loss function value; the smaller the degree of difference between the user predicted travel preference and the pre-collected user travel preference tags, the smaller the first loss function value.
That is, in the training process of the user model, the functions of applying the historical behavior data and the historical preference data and predicting the travel preference of the user are learned.
Here, the parameters of the initial user model may be updated using a cross entropy loss function, a mean square error loss function, or a random gradient descent method, which is not particularly limited in the embodiment of the present invention.
After training to obtain the user model, the user model can be trained again to obtain an initial travel recommendation model.
In this process, sample preference data, sample demand data, and user travel demand labels of the user may be collected in advance.
After the user model is obtained, sample preference data, sample demand data and user travel demand labels of the user which are collected in advance can be applied to train the user model:
first, sample preference data and sample demand data of a user can be input into a user model, and a user model can obtain and output a user predicted travel demand.
After obtaining the user predicted travel demand based on the user model, the user predicted travel demand can be compared with the user travel demand label collected in advance, a second loss function value is calculated according to the difference degree between the user predicted travel demand and the user travel demand label, parameter iteration is carried out on the user model based on the second loss function value, and the user model after parameter iteration is completed is recorded as an initial travel recommendation model.
It will be appreciated that the greater the degree of difference between the user's predicted travel demand and the pre-collected user travel demand labels, the greater the second loss function value; the smaller the degree of difference between the user's predicted travel demand and the pre-collected user travel demand labels, the smaller the second loss function value.
Here, the cross entropy loss function, the mean square error loss function, and the random gradient descent method may be used to update the parameters of the user model, which is not particularly limited in the embodiment of the present invention.
It should be noted that the specific process of travel recommendation is as follows:
and (3) personalized analysis: through deep learning techniques, the system analyzes the user's historical behavior, preferences, features, and set travel times and places to build a highly personalized user model. Destination feature recommendation integration: the system obtains detailed information about travel destinations, including cultures, delicacies, attractions, etc., to generate suggestions based on the user's preferences and destination characteristics. And (3) real-time data collection: the system collects traffic conditions, spot warmth, queuing time and congestion information about the day in real time. Travel advice prediction: deep learning models, i.e., systems that combine user models, destination recommendation data, and real-time data for tourist attractions, are utilized to predict attractions, dining venues, and cultural experiences that may be of interest to a user. Comprehensive advice: finally, the system generates highly personalized travel recommendations, including attraction recommendations, dining recommendations, cultural experiences, and provides reference information about traffic conditions, attraction warmth, queuing times, and congestion.
Therefore, the problem that the traditional travel strategy generation method cannot provide comprehensive information about scenic spot queuing conditions, congestion and real-time traffic conditions is solved, and the comprehensiveness and accuracy of a traveler in planning a journey are improved.
Based on the above embodiment, the step of obtaining the sample preference data and the sample demand data of the user includes:
step 310, obtaining the sample historical behavior data and the sample historical preference data;
step 320, extracting the sample preference data and the sample demand data from the sample historical behavior data and the sample historical preference data.
Specifically, it is considered that the sample preference data and sample demand data of the user are associated with the sample historical behavior data and sample historical preference data of the user in a large manner.
Accordingly, sample historical behavior data and sample historical preference data of the user can be obtained, and then sample preference data and sample demand data are extracted from the sample historical behavior data and the sample historical preference data of the user.
Here, the sample historical behavior data of the user may be a place where the user has visited, a selected traffic mode, a stay time, etc., and these information may provide insight about travel preference and habit of the user, or may be actions such as praise, collection, scoring, etc. of the user in the application program, which is not limited in particular by the embodiment of the present invention.
The sample historical preference data may include activities, restaurants, attractions, etc. that the user likes, as embodiments of the invention are not specifically limited.
The sample demand data is used to reflect personal preferences, health conditions, eating habits, etc. of the user during the historical travel, and embodiments of the invention are not specifically limited in this regard.
Based on the above embodiment, step 130 further includes:
step 131, obtaining feedback data of a user;
and step 132, carrying out parameter adjustment on the travel recommendation model based on the feedback data of the user.
In particular, after generating the travel advice based on the travel recommendation model, the user may provide an assessment of the travel advice, actual experience feedback for the travel advice, and any special demand changes, given that the user may be dissatisfied or confused. Thus, feedback data of the user can be acquired.
The evaluation of the travel advice, the actual experience feedback of the travel advice and any special requirement change are considered to relate to the privacy security of the user, so that after the feedback data of the user are acquired, the privacy processing can be carried out on the feedback data of the user, the privacy of the user is prevented from being revealed, and the use experience of the user is reduced.
Here, the feedback data of the user is used to reflect the opinion of the user about the travel suggestion, and the feedback data of the user may be in the form of interactive voice or interactive text, which is not limited in particular by the embodiment of the present invention.
Under the condition that the feedback data of the user is interactive voice, the interactive voice of the user can be subjected to voice recognition to obtain an interactive text, and then parameter adjustment is performed on the travel recommendation model based on the interactive text.
Here, the interactive voice may be obtained through a sound pickup device, where the sound pickup device may be an intelligent camera, or may be an intelligent sound, or may be a notebook, or the like, and the sound pickup device may amplify and reduce noise of the interactive voice after obtaining the interactive voice through the microphone array pickup.
Here, the HMM (Hidden Markov Model ), deep model, waveNet, etc. may be used for performing the voice recognition on the interactive voice of the user, which is not particularly limited in the embodiment of the present invention.
According to the method provided by the embodiment of the invention, the feedback data of the user is obtained, and the parameter adjustment is performed on the travel recommendation model based on the feedback data of the user, so that the generated suggestion is continuously improved, the accuracy is improved, and the user requirements are met.
Based on the above embodiment, step 132 includes:
step 1321, determining the frequency and depth of the interaction topics in the feedback data of the user;
step 1322, performing parameter adjustment on the travel recommendation model based on the interaction topic frequency and the interaction topic depth.
Specifically, the frequency of the interaction topics and the depth of the interaction topics in the feedback data of the user can be determined, wherein the frequency of the interaction topics reflects the feedback times of the user based on the travel advice, and the depth of the interaction topics can be determined by using the interaction times of the user on the same travel advice.
After determining the frequency and depth of the interaction topics in the feedback data of the user, parameter adjustment can be performed on the travel recommendation model based on the frequency and depth of the interaction topics.
It can be appreciated that the more frequently the topics are interacted, the better the travel advice generated by the travel recommendation model; the less frequently the topics are interacted, the worse the travel advice generated by the travel recommendation model.
It can be appreciated that the deeper the depth of the interaction topic, the better the travel advice generated by the travel recommendation model; the shallower the depth of the interaction topic, the worse the travel advice generated by the travel recommendation model.
In addition, the average duration of the interaction topics in the feedback data of the user can be combined to carry out parameter adjustment on the travel recommendation model, and the embodiment of the invention is not particularly limited.
Based on any of the above embodiments, fig. 2 is a second flow chart of a travel recommendation method according to the present invention, as shown in fig. 2, the method includes:
first, the user can set travel time, location and special needs in the application, such as home friendly, cultural exploration, travel on food, etc.
And secondly, acquiring sample historical behavior data and sample historical preference data, and extracting sample preference data and sample demand data from the sample historical behavior data and the sample historical preference data.
And thirdly, training a user model based on the sample historical behavior data, the sample historical preference data and the user travel preference labels of the user.
Fourth, acquiring real-time data of tourist attractions and destination recommendation data. The real-time data of the tourist attractions comprise real-time traffic conditions, real-time attraction heat, real-time queuing time length and real-time congestion information. These data will be used to generate real-time suggestions. The real-time traffic condition data includes public traffic conditions, road congestion conditions, and the like. The real-time scenic spot heat data reflects the real-time heat of the scenic spot, and is based on the number of tourists, evaluation and other data. The real-time queuing time length information provides the real-time queuing condition of the scenic spot. The real-time congestion information indicates congestion areas and conditions in the city.
Destination recommendation data comprises cultures, delicacies, scenic spots and the like, and the information is used as the basis for generating travel advice. More specifically, the destination recommendation data may include history of the destination, cultural heritage, famous attractions, featured restaurants, etc.
Fifthly, real-time data and destination recommendation data of tourist attractions can be input into a tourist recommendation model, and tourist suggestions are obtained and output by the tourist recommendation model; the first step user can also set travel time, position and special requirements in the application program, such as family friendly, cultural exploration, food travel and the like, and the real-time data of tourist attractions and destination recommendation data are input into the tourist recommendation model together, and the tourist recommendation model obtains and outputs the tourist recommendation.
And sixthly, acquiring feedback data of the user, and determining the frequency and depth of the interaction topics in the feedback data of the user.
And carrying out parameter adjustment on the travel recommendation model based on the interaction topic frequency and the interaction topic depth.
The training step of the travel recommendation model comprises the following steps: acquiring an initial travel recommendation model, sample real-time data of tourist attractions, sample destination recommendation data and tag travel suggestions;
inputting sample real-time data and sample destination recommendation data into an initial travel recommendation model, and obtaining and outputting predicted travel suggestions by the initial travel recommendation model;
and determining a loss function value based on the difference between the predicted travel suggestion and the labeled travel suggestion, and performing parameter iteration on the initial travel recommendation model based on the loss function value to obtain the travel recommendation model.
The training step of the initial travel recommendation model comprises the following steps:
training a user model based on the user's sample historical behavior data, sample historical preference data, and user travel preference tags;
based on the sample preference data and sample demand data of the user and the travel demand labels of the user, training the user model to obtain an initial travel recommendation model.
The travel recommendation device provided by the invention is described below, and the travel recommendation device described below and the travel recommendation method described above can be correspondingly referred to each other.
Based on any one of the above embodiments, the present invention provides a travel recommendation device, and fig. 3 is a schematic structural diagram of the travel recommendation device provided by the present invention, as shown in fig. 3, the device includes:
an acquiring unit 310, configured to acquire real-time data of a tourist attraction and destination recommendation data;
a travel advice unit 320, configured to input real-time data of the tourist attraction and the destination recommendation data into a travel recommendation model, obtain and output travel advice from the travel recommendation model;
the travel recommendation model is obtained based on sample real-time data of tourist attractions, sample destination recommendation data and label travel advice training.
The device provided by the embodiment of the invention acquires the real-time data of the tourist attractions and the destination recommendation data, inputs the real-time data of the tourist attractions and the destination recommendation data into the tourist recommendation model, and obtains and outputs the tourist advice by the tourist recommendation model, wherein the tourist recommendation model is obtained based on the sample real-time data of the tourist attractions, the sample destination recommendation data and the label tourist advice training. On one hand, according to different travel destinations, destination recommendation data are different, and then travel recommendation is performed based on the destination recommendation data, so that individuation and accuracy of travel recommendation are improved; on the other hand, the real-time data of tourist attractions is referenced, so that the experience of travelers is optimized, more accurate travel advice is provided, the travelers are helped to plan the journey better, queuing and congestion are avoided, and the specific requirements and preferences of the travelers are fully met.
Based on any one of the above embodiments, the obtaining step of the travel recommendation model includes:
acquiring an initial travel recommendation model, sample real-time data of the tourist attraction, sample destination recommendation data and the label travel suggestion;
inputting the sample real-time data and the sample destination recommendation data into the initial travel recommendation model, and obtaining and outputting predicted travel suggestions by the initial travel recommendation model;
and determining a loss function value based on the difference between the predicted travel advice and the tag travel advice, and carrying out parameter iteration on the initial travel recommendation model based on the loss function value to obtain the travel recommendation model.
Based on any of the above embodiments, the training step of the initial travel recommendation model includes:
training a user model based on the user's sample historical behavior data, sample historical preference data, and user travel preference tags;
training the user model based on sample preference data and sample demand data of the user and a user travel demand label to obtain the initial travel recommendation model.
Based on any of the above embodiments, the step of obtaining the sample preference data and the sample demand data of the user includes:
acquiring the sample historical behavior data and the sample historical preference data;
and extracting the sample preference data and the sample demand data from the sample history behavior data and the sample history preference data.
Based on any one of the above embodiments, the system further includes a user feedback unit, where the user feedback unit specifically includes:
the method comprises the steps of acquiring a feedback data unit, which is used for acquiring feedback data of a user;
and the parameter adjustment unit is used for adjusting parameters of the travel recommendation model based on the feedback data of the user.
Based on any of the above embodiments, the parameter adjustment unit is specifically configured to:
determining the frequency and depth of the interaction topics in the feedback data of the user;
and carrying out parameter adjustment on the travel recommendation model based on the interaction topic frequency and the interaction topic depth.
Fig. 4 illustrates a physical schematic diagram of an electronic device, as shown in fig. 4, which may include: processor 410, communication interface (Communications Interface) 420, memory 430 and communication bus 440, wherein processor 410, communication interface 420 and memory 430 communicate with each other via communication bus 440. The processor 410 may invoke logic instructions in the memory 430 to perform a travel recommendation method comprising: acquiring real-time data of tourist attractions and destination recommendation data; inputting the real-time data of the tourist attractions and the destination recommendation data into a tourist recommendation model, and obtaining and outputting tourist suggestions by the tourist recommendation model; the travel recommendation model is obtained based on sample real-time data of tourist attractions, sample destination recommendation data and label travel advice training.
Further, the logic instructions in the memory 430 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program, the computer program being storable on a non-transitory computer readable storage medium, the computer program, when executed by a processor, being capable of performing the travel recommendation method provided by the methods described above, the method comprising: acquiring real-time data of tourist attractions and destination recommendation data; inputting the real-time data of the tourist attractions and the destination recommendation data into a tourist recommendation model, and obtaining and outputting tourist suggestions by the tourist recommendation model; the travel recommendation model is obtained based on sample real-time data of tourist attractions, sample destination recommendation data and label travel advice training.
In yet another aspect, the present invention provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the travel recommendation method provided by the methods described above, the method comprising: acquiring real-time data of tourist attractions and destination recommendation data; inputting the real-time data of the tourist attractions and the destination recommendation data into a tourist recommendation model, and obtaining and outputting tourist suggestions by the tourist recommendation model; the travel recommendation model is obtained based on sample real-time data of tourist attractions, sample destination recommendation data and label travel advice training.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A travel recommendation method, comprising:
acquiring real-time data of tourist attractions and destination recommendation data;
inputting the real-time data of the tourist attractions and the destination recommendation data into a tourist recommendation model, and obtaining and outputting tourist suggestions by the tourist recommendation model;
the travel recommendation model is obtained based on sample real-time data of tourist attractions, sample destination recommendation data and label travel advice training.
2. The travel recommendation method according to claim 1, wherein the obtaining step of the travel recommendation model includes:
acquiring an initial travel recommendation model, sample real-time data of the tourist attraction, sample destination recommendation data and the label travel suggestion;
inputting the sample real-time data and the sample destination recommendation data into the initial travel recommendation model, and obtaining and outputting predicted travel suggestions by the initial travel recommendation model;
and determining a loss function value based on the difference between the predicted travel advice and the tag travel advice, and carrying out parameter iteration on the initial travel recommendation model based on the loss function value to obtain the travel recommendation model.
3. The travel recommendation method according to claim 2, wherein the training step of the initial travel recommendation model comprises:
training a user model based on the user's sample historical behavior data, sample historical preference data, and user travel preference tags;
training the user model based on sample preference data and sample demand data of the user and a user travel demand label to obtain the initial travel recommendation model.
4. The travel recommendation method according to claim 3, wherein the step of obtaining sample preference data and the sample demand data of the user comprises:
acquiring the sample historical behavior data and the sample historical preference data;
and extracting the sample preference data and the sample demand data from the sample history behavior data and the sample history preference data.
5. The travel recommendation method according to any one of claims 1 to 4, wherein said inputting the real-time data of the tourist attraction and the destination recommendation data into a travel recommendation model, obtaining and outputting travel advice from the travel recommendation model, further comprises:
acquiring feedback data of a user;
and carrying out parameter adjustment on the travel recommendation model based on the feedback data of the user.
6. The travel recommendation method according to claim 5, wherein said parameter adjustment of said travel recommendation model based on said user feedback data comprises:
determining the frequency and depth of the interaction topics in the feedback data of the user;
and carrying out parameter adjustment on the travel recommendation model based on the interaction topic frequency and the interaction topic depth.
7. A travel recommendation device, comprising:
the acquisition unit is used for acquiring real-time data of tourist attractions and destination recommendation data;
the tourist recommendation unit is used for inputting the real-time data of the tourist attraction and the destination recommendation data into a tourist recommendation model, and obtaining and outputting tourist recommendation by the tourist recommendation model;
the travel recommendation model is obtained based on sample real-time data of tourist attractions, sample destination recommendation data and label travel advice training.
8. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the travel recommendation method of any one of claims 1 to 6 when the program is executed by the processor.
9. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements the travel recommendation method according to any one of claims 1 to 6.
10. A computer program product comprising a computer program which, when executed by a processor, implements the travel recommendation method as claimed in any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311618297.3A CN117540098A (en) | 2023-11-29 | 2023-11-29 | Travel recommendation method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311618297.3A CN117540098A (en) | 2023-11-29 | 2023-11-29 | Travel recommendation method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117540098A true CN117540098A (en) | 2024-02-09 |
Family
ID=89784047
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311618297.3A Withdrawn CN117540098A (en) | 2023-11-29 | 2023-11-29 | Travel recommendation method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117540098A (en) |
-
2023
- 2023-11-29 CN CN202311618297.3A patent/CN117540098A/en not_active Withdrawn
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sperlí | A cultural heritage framework using a Deep Learning based Chatbot for supporting tourist journey | |
| JP6753707B2 (en) | Artificial intelligence system that supports communication | |
| US10635707B2 (en) | Contextual memory bandit for proactive dialogs | |
| Campigotto et al. | Personalized and situation-aware multimodal route recommendations: the FAVOUR algorithm | |
| US20100217730A1 (en) | Temporally-controlled item recommendation method and system based on rating prediction | |
| Edelmann et al. | Cultural resources and cultural distinction in networks | |
| US11205128B2 (en) | Inferred profiles on online social networking systems using network graphs | |
| CN105453606B (en) | Air navigation aid and navigation equipment | |
| Khatibi et al. | Fine-grained tourism prediction: Impact of social and environmental features | |
| US20170293860A1 (en) | System and methods for suggesting beneficial actions | |
| CN107330130A (en) | A kind of implementation method of dialogue robot to artificial customer service recommendation reply content | |
| US12073323B2 (en) | System and method for intelligent service intermediation | |
| CN114398556A (en) | Learning content recommendation method, device, equipment and storage medium | |
| Thiengburanathum et al. | A decision tree based recommendation system for tourists | |
| Anand et al. | Chatbot enabled smart tourism service for indian cities: An ai approach | |
| Sai et al. | A Web-Based Chatbot for Indian Cities: A Comparison of CNN, ANN, and LSTM Models | |
| Gehlot et al. | Research guide for ML based smart tourist system | |
| CN112784177A (en) | Spatial distance adaptive next interest point recommendation method | |
| CN117540098A (en) | Travel recommendation method and device, electronic equipment and storage medium | |
| US20230325944A1 (en) | Adaptive wellness collaborative media system | |
| Dareddy | Challenges in Recommender Systems for Tourism. | |
| Kelleher et al. | Finding common ground for citizen empowerment in the smart city | |
| Widener | Populist Placemaking: Grounds for Open Government-Citizen Spatial Regulating Discourse | |
| US9532165B2 (en) | Method and apparatus for location prediction using short text | |
| Ngamsa-Ard et al. | Point-of-interest (POI) recommender systems for social groups in location based social networks (LBSNs): Proposition of an improved model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20240209 |