Information recommendation method, device, equipment and computer storage medium

Abstract

This invention relates to the field of computer data processing technology and discloses an information recommendation method. The method includes: determining the historical travel trajectory data of a target user; the historical travel trajectory data corresponding to multiple historical tourist attractions; determining the target user's patience level based on the historical travel trajectory data; the patience level characterizes the degree to which the number of times the target user has visited historical tourist attractions affects the target user's interest in those attractions; determining the target user's predicted interest level in each of the predicted attractions based on the patience level, the attraction's feature information, and the attraction's feature information; and determining the target recommended attraction corresponding to the target user based on the predicted interest level of each of the predicted attractions. Through this method, this invention improves the accuracy of tourist attraction recommendations.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of computer data processing technology, specifically to an information recommendation method, apparatus, device, and computer storage medium. Background Technology

[0002] Information recommendation is a common user need across various fields. For tourist attraction recommendations, existing technologies typically use collaborative filtering based on the similarity between the attraction to be recommended and the user's historical tourist attractions. However, the inventors of this application discovered during the implementation of the embodiments of this application that existing tourist attraction recommendations have low accuracy. Summary of the Invention

[0003] In view of the above problems, embodiments of the present invention provide an information recommendation method to solve the problem of low accuracy in tourist attraction recommendations in the prior art.

[0004] According to one aspect of the present invention, an information recommendation method is provided, the method comprising:

[0005] Determine the target user's historical travel trajectory data; the historical travel trajectory data corresponds to multiple historical tourist attractions;

[0006] The patience level of the target user is determined based on the historical travel trajectory data; the patience level is used to characterize the degree to which the number of times the target user visits the historical tourist attractions affects the target user's interest in the historical tourist attractions.

[0007] The target user's predicted interest in each of the predicted attractions is determined based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each attraction to be predicted.

[0008] The target recommended attractions for the target user are determined based on the predicted interest level of each of the attractions to be predicted.

[0009] In one optional approach, the historical tourism trajectory data includes the attraction feature information and tourism time information of each of the historical tourist attractions; the method further includes:

[0010] The historical tourist attractions are classified into at least one category based on their characteristic information; each category includes at least one of the historical tourist attractions.

[0011] Based on the travel time information, determine the target user's historical travel frequency for each of the aforementioned attraction categories;

[0012] Determine the initial level of interest for each of the aforementioned historical tourist attractions;

[0013] The total interest level of the historical tourism trajectory data is determined based on the initial interest level of each historical tourist attraction, the number of historical visits to the attraction category, and the initial patience level of the target user.

[0014] The initial patience level is optimized with the goal of maximizing the total interest value, and the optimal solution is determined as the patience level.

[0015] In an alternative approach, the method further includes:

[0016] The actual level of interest in each of the aforementioned historical tourist attractions is determined according to the following formula;

[0017]

[0018] Among them, I p Let k+1 be any one of the historical tourist attractions mentioned above, where k+1 represents I. p The (k+1)th of all the historical tourist attractions mentioned was visited by the target user. For I p The actual level of interest, E p 'For I p The initial degree of interest, For I p The historical number of visits to the attraction category, where σ' represents the initial patience level;

[0019] The total interest score is determined based on the actual interest scores of all the historical tourist attractions in the historical travel trajectory data.

[0020] In an alternative approach, the method further includes:

[0021] Multiple particles and their corresponding initial velocity vectors are randomly generated, wherein each particle corresponds to a patience value.

[0022] Based on the patience level corresponding to each particle, the total interest level corresponding to each particle is determined.

[0023] The total interest is determined as a reference value for each of the particles;

[0024] Based on the reference values, determine the local optimal solution corresponding to each particle and the global optimal solution corresponding to all particles;

[0025] The initial velocity vectors of each particle are updated based on the local optimal solution and the global optimal solution to obtain the updated velocity vectors of each particle.

[0026] The updated position of each particle and the patience level corresponding to the updated position are determined based on the current position of each particle and the updated velocity vector.

[0027] Determine whether the updated vector velocity of each particle meets the vector velocity threshold;

[0028] When the vector velocity threshold is satisfied, the global optimal solution is determined as the optimal solution.

[0029] In an alternative approach, the method further includes:

[0030] The similarity between the historical tourist attraction and the attraction to be predicted is determined based on the attraction feature information of the historical tourist attraction and the attraction to be predicted.

[0031] The predicted number of visits to the destination is determined based on the similarity.

[0032] Determine the initial level of interest in the tourist attraction to be predicted;

[0033] The predicted interest level of the destination is determined based on the patience level, the predicted number of trips, and the initial interest level of the destination.

[0034] In an alternative approach, the method further includes:

[0035] Determine the sum of the similarities between the proposed attraction and all the historical tourist attractions;

[0036] The predicted number of trips is determined based on the sum of the similarities.

[0037] In an alternative approach, the method further includes:

[0038] The predicted interest level of the scenic spot to be predicted is determined according to the following formula:

[0039]

[0040] Where E is the predicted interest level of the attraction to be predicted, E' is the initial interest level of the attraction to be predicted, t is the predicted number of trips, and σ is the patience level.

[0041] According to another aspect of the present invention, an information recommendation device is provided, comprising:

[0042] The first determining module is used to determine the target user's historical travel trajectory data; the historical travel trajectory data corresponds to multiple historical tourist attractions.

[0043] The second determining module is used to determine the patience level of the target user based on the historical travel trajectory data; the patience level is used to characterize the degree of influence of the number of times the target user visits the historical tourist attractions on the target user's interest in the historical tourist attractions;

[0044] The prediction module is used to determine the target user's predicted interest in each of the predicted attractions based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each attraction to be predicted.

[0045] The recommendation module is used to determine the target recommended attractions corresponding to the target user based on the predicted interest level of each of the attractions to be predicted.

[0046] According to another aspect of the present invention, an information recommendation device is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus;

[0047] The memory is used to store at least one executable instruction that causes the processor to perform operations as described in any of the information recommendation method embodiments.

[0048] According to another aspect of the present invention, a computer-readable storage medium is provided, the storage medium storing at least one executable instruction that causes an information recommendation device to perform the operation of any of the information recommendation methods described herein.

[0049] This invention addresses the issue of low accuracy in tourist attraction recommendations caused by collaborative filtering in existing technologies. By extracting historical travel trajectory data (corresponding to multiple historical tourist attractions) from the user's historical travel trajectory data, this invention improves the accuracy of tourist attraction recommendations by integrating this data with collaborative filtering. This improves the accuracy of tourist attraction recommendations by mining the user's historical travel trajectory data to understand the influence of the number of visits to historical tourist attractions on the user's interest in those attractions. The patience level represents the degree to which the user's interest in a particular historical tourist attraction decreases as the number of visits increases. The lower the patience level, the greater the decrease in interest as the user visits a particular historical tourist attraction more frequently. The invention also determines the predicted interest level of each potential attraction based on the patience level, the number of visits to historical tourist attractions, and the predicted interest level of each potential attraction.

[0050] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0051] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0052] Figure 1 A flowchart illustrating the information recommendation method provided in an embodiment of the present invention is shown;

[0053] Figure 2 A schematic diagram of the information recommendation device provided in an embodiment of the present invention is shown;

[0054] Figure 3 A schematic diagram of the information recommendation device provided in an embodiment of the present invention is shown. Detailed Implementation

[0055] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0056] Figure 1 A flowchart of an information recommendation method provided by an embodiment of the present invention is shown, the method being executed by a computer processing device. This computer processing device may include a mobile phone, a laptop computer, etc. Figure 1 As shown, the method includes the following steps:

[0057] Step 10: Determine the target user's historical travel trajectory data; the historical travel trajectory data corresponds to multiple historical tourist attractions.

[0058] In one embodiment of the present invention, historical travel trajectory data can correspond to related data of a sequence of historical tourist attractions. The sequence of historical tourist attractions records multiple historical tourist attractions visited by the target user in chronological order of their travel time. The division of the historical tourist attraction sequence can be determined based on the time interval between the user's visits to each historical tourist attraction. When the time interval exceeds a preset threshold, it is considered the end of a segment of the historical travel trajectory.

[0059] The relevant data for the historical tourist attraction sequence can include the attraction characteristics and travel time information of each historical tourist attraction. The travel time information can include the absolute travel time of the target user's visit to the historical tourist attraction, such as 10:00:00 on January 1, 2021, and the relative travel time of the historical tourist attraction, such as the 3rd historical tourist attraction out of a total of 10 historical tourist attractions in the sequence.

[0060] Step 20: Determine the patience level of the target user based on the historical travel trajectory data.

[0061] According to the theory of marginal utility in economics, the more times a user visits similar attractions, the more the current effect of those attractions on the user will be affected by marginal utility, resulting in a lower actual effect. The effect of an attraction on a user refers to the benefits the user gains from visiting that attraction. Assuming the user is a rational individual, the greater this effect, the greater the user's interest in the attraction; that is, interest and effect can be considered equivalent mappings.

[0062] Therefore, in one embodiment of the present invention, by setting a patience level, the degree of this effect reduction can be controlled, thereby simulating the user's boredom and obtaining a more realistic effect value, resulting in a more accurate recommendation. The patience level is used to characterize the influence of the target user's number of visits to the historical tourist attraction on the target user's interest in the historical tourist attraction. That is, the greater the patience level, the slower the effect of the attraction on the user decreases as the number of visits increases, meaning the user has the willingness to visit the tourist attraction several times; conversely, the user's effect decreases faster as the number of visits increases, and the user is more likely to become bored with the tourist attraction. Furthermore, when the number of visits is less than the user's patience level, the effect of the attraction on the user decreases less as the number of visits increases, and the effect of the attraction on the user is also higher; conversely, the effect of the attraction on the user decreases rapidly as the number of visits increases, eventually approaching 0.

[0063] In another embodiment of the invention, the historical travel trajectory data includes multiple historical tourist attractions arranged in the order of travel. The patience level can be optimized by taking the maximum sum of the effects of each attraction on the user at that time as the optimization objective, and then optimizing a preset initial patience level. Since it can be assumed that the user is intelligent and wise, during the formation of the user's historical travel trajectory, each travel choice will select the attraction that has the greatest effect on them. Therefore, this overall travel trajectory effect is the target value of the patience factor, and the optimal value of the patience factor is the value that maximizes the overall effect of the user's travel trajectory.

[0064] In one embodiment of the present invention, the effect of a user on each historical tourist attraction can be obtained by weighting the initial effect of the attraction based on the number of times the user has visited the historical tourist attraction and the user's current level of patience.

[0065] In another embodiment of the present invention, cluster analysis can be performed on historical tourist attractions in historical travel trajectories. Multiple historical tourist attractions are divided into multiple attraction categories according to the distance of their attraction feature information. Then, the total number of historical visits to each attraction category by the target user is taken as the number of historical visits to that attraction.

[0066] In another embodiment of the present invention, when weighting the initial effect based on the number of historical trips and the initial patience level, the number of historical trips can be determined as the abscissa of the peak value of the Gaussian function, and the initial patience level can be used as the standard deviation of the Gaussian function.

[0067] Therefore, in another embodiment of the present invention, the historical tourism trajectory data includes the attraction feature information and tourism time information of each of the historical tourist attractions. Step 20 further includes:

[0068] Step 201: Divide the historical tourist attractions into at least one category based on their attraction characteristics; each category includes at least one of the historical tourist attractions.

[0069] In one embodiment of the present invention, the distance between various historical tourist attractions is calculated based on the attraction feature information, and the historical tourist attractions are clustered based on the distance to obtain at least one attraction class.

[0070] Step 202: Determine the number of times the target user has visited each of the aforementioned attractions based on the travel time information.

[0071] In one embodiment of the present invention, the number of visits to each historical tourist attraction is counted based on tourism time information, and the sum of all the numbers in each attraction category is determined as the historical number of visits corresponding to that attraction category.

[0072] Step 203: Determine the initial level of interest for each of the aforementioned historical tourist attractions.

[0073] In one embodiment of the present invention, collaborative filtering can be used to determine the initial interest level of historical tourist attractions by the distance between dimension values ​​under the same dimension. The dimension of collaborative filtering can be the attraction's feature information or the user feature information of users who have visited the attraction.

[0074] In another embodiment of the present invention, when the dimension of collaborative filtering is attraction feature information, it can be based on the principle that if most users who have visited attraction A have also visited attraction B, then attraction A and attraction B are similar. For two attractions A and B, the similarity between the two attractions can be expressed as:

[0075]

[0076] Where N(A∩B) is the set of users who have visited both attraction A and attraction B; N(*) is the set of all users who have visited attraction *.

[0077] In another embodiment of the present invention, when the dimension of collaborative filtering is user feature information, the calculation of similarity between users can be based on the following assumptions: the more users with the same attributes, the higher the similarity; or, for users who have visited more similar tourist attractions, the higher the similarity between them.

[0078] Therefore, user similarity can be defined by user attributes. For two users A and B, if user A has a tag set L A ={l A,1 ,l A,2 ,...l A,m User B has a tag set L. B ={l B,1 ,l B,2 ,...l B,n The similarity between user A and user B is defined as follows:

[0079]

[0080] In another embodiment of the present invention, for a user, if user A has a travel trajectory T A =(I1,I2,...,I m If user A has a feature set F, then user A has a feature set F. A ={I1,I2,...,I m}

[0081] For two feature sets F1 and F2, if there are The feature sets are then transformed into feature vectors in sequence. The similarity between two feature sets is the modulus of the vector composed of the maximum similarity between each dimension of the feature vector transformed by F2 and any feature in F1. This can be expressed by the first formula as follows:

[0082]

[0083] Where m and n are the moduli of feature sets F1 and F2, respectively.

[0084] For two feature sets F1 and F2, F1∩2 The similarity between two feature sets is defined as the feature vector generated after removing duplicate options from two feature sets according to a specific sorting rule. with vector The cosine similarity between them can be expressed by the second formula as follows:

[0085]

[0086] If the similarity between any two attractions is known, then the similarity between the travel trajectories of any two users can be determined according to the first and second formulas mentioned above, thereby defining the similarity between users.

[0087] The similarity between users can be obtained from the two definitions of user similarity as follows:

[0088]

[0089] Where α is a parameter related to the length of the travel trajectories of the two users; α = exp(-min(||F A ||,||F B ||)).

[0090] Therefore, when recommending attractions based on collaborative filtering, recommendations can be made based on attractions visited by similar users, or based on similar attractions visited by the user. Specifically, the interest level calculation based on attractions visited by similar users is as follows: for a given user and a given attraction, the user's interest level in that attraction is the average similarity between the user and other users who have visited that attraction, as follows:

[0091]

[0092] Where N(C) is the set of all users who have visited attraction C.

[0093] In another embodiment of the present invention, the initial interest level based on visited tourist attractions is calculated as follows: for a user and a tourist attraction, the user's interest level in the tourist attraction is the maximum similarity between the user and all other tourist attractions visited by that user.

[0094]

[0095] Finally, the initial level of user interest in the attraction is defined as:

[0096] P(A,C)=max(P U (A,C),P I (A,C)); This gives the user's initial interest in all tourist attractions.

[0097] In another embodiment of the present invention, when calculating the similarity between users and / or attractions, the cold start problem of the tourist attraction recommendation system is taken into consideration. This problem is caused by insufficient user data in the initial stage of system operation, which makes it impossible to establish the corresponding similarity model, thus causing the recommendation algorithm to fail to operate normally.

[0098] Therefore, for calculating user similarity during cold starts, we can obtain the user's relevant information and favorite attraction tags entered during the registration or login process as their corresponding initial preferences. For a given user, each attribute of the user and each favorite attraction tag entered by the user determines a feature tag for that user.

[0099] To calculate the similarity of attractions during a cold start, each attraction can be assigned a corresponding tag, which serves as the attraction's attribute set. Specifically, for two objects A and B, if object A has a tag set L... A ={l A,1 ,l A,2 ,...,l A,m Object B has a tag set L. B ={l B,1 ,l B,2 ,...,l B,n The similarity between object A and object B is defined as follows:

[0100]

[0101] Based on the above definition of similarity, the similarity of all attractions and users can be obtained without data accumulation. When there is no user trajectory record, i.e. no corresponding boundary effect model, the recommendation result of the collaborative filtering recommendation algorithm can be used directly. When there is no corresponding record of similar users or the user itself, non-personalized recommendation results can be used for direct recommendation.

[0102] Step 204: Determine the total interest of the historical travel trajectory data based on the initial interest of each historical tourist attraction, the number of historical visits to the attraction category, and the initial patience of the target user.

[0103] In one embodiment of the present invention, a weighted influence factor can be determined based on the Gaussian function and the initial interest level, according to the historical number of visits and the initial patience level. The product of the weighted influence factor and the initial interest level is taken as the actual interest level of the historical tourist attractions. Finally, the sum of the actual interest levels of each historical tourist attraction included in the historical tourist trajectory data is determined as the total interest level of the historical tourist trajectory data.

[0104] Therefore, in another embodiment of the present invention, step 204 further includes:

[0105] Step 2041: Determine the actual interest level of each of the aforementioned historical tourist attractions according to the following formula:

[0106]

[0107] Among them, I p Let k+1 be any one of the historical tourist attractions mentioned above, where k+1 represents I. p The (k+1)th of all the historical tourist attractions mentioned was visited by the target user. For I p The actual level of interest, E p 'For I p The initial degree of interest, For I p The historical number of visits to the attraction category, σ' is the initial patience level.

[0108] In one embodiment of the present invention, according to the calculation formula of actual interest, the sign of patience is not related to the effect value of the attraction. Therefore, the theoretical range of patience can be defined as [0, +∞). Given that the initial effect of each tourist attraction on the user and the similarity between tourist attractions are determined, the problem is transformed into finding the value of patience that maximizes the overall effect of the user's travel trajectory within the solution space of patience.

[0109] In another embodiment of the present invention, it is known from the survey of users’ patience that the actual range of patience is not that large, and is generally concentrated in the range of [2,5]. Therefore, in order to improve the applicability of the marginal effect calculation model of interest and meet the needs of some people with special preferences, the range of patience is set to (0,20], and the value of patience that maximizes the overall effect of the user’s travel trajectory is found within this range.

[0110] Step 2042: Determine the total interest score based on the actual interest scores of all the historical tourist attractions in the historical travel trajectory data.

[0111] Step 205: Optimize the initial patience level with the goal of maximizing the total interest value, and determine the optimal solution as the patience level.

[0112] In one embodiment of the present invention, when optimizing the initial patience level, heuristic algorithms such as simulated annealing, genetic algorithms, and ant colony algorithms can be used to speed up the optimization process.

[0113] In another embodiment of the present invention, a particle swarm optimization algorithm can be used to solve for the optimal solution of patience, and step 205 further includes:

[0114] Step 2051: Randomly generate multiple particles and their corresponding initial velocity vectors, wherein each particle corresponds to a patience value.

[0115] In one embodiment of the present invention, the patience value corresponding to each particle is randomly generated.

[0116] Step 2052: Determine the total interest degree corresponding to each particle based on the patience degree corresponding to each particle.

[0117] In one embodiment of the present invention, the total interest degree corresponding to each particle is calculated according to steps 2041 and 2042.

[0118] Step 2053: Determine the total interest as the reference value for each of the particles.

[0119] Step 2054: Determine the local optimal solution corresponding to each particle and the global optimal solution corresponding to all particles based on the reference value.

[0120] In one embodiment of the present invention, the maximum reference value within a preset range of each particle is determined as the local optimal solution corresponding to that particle, and the maximum reference value among all particles is determined as the global optimal solution.

[0121] Step 2055: Update each of the initial velocity vectors according to the local optimal solution and the global optimal solution to obtain the updated velocity vectors of each particle.

[0122] In one embodiment of the present invention, the initial velocity vector can be updated according to the acceleration constant and inertia factor set in the particle swarm algorithm, wherein the acceleration constant is used to control the weights of individual experience and swarm experience in the velocity vector update, and the inertia factor is used to control the number of particles that inherit the current velocity.

[0123] Step 2056: Determine the updated position of each particle and the patience level corresponding to the updated position based on the current position of each particle and the updated velocity vector.

[0124] In one embodiment of the present invention, starting from the current position, the updated position is calculated based on the updated velocity vector, and the reference value corresponding to the updated position is determined as the corresponding patience level.

[0125] Step 2057: Determine whether the updated vector velocity of each particle meets the vector velocity threshold.

[0126] In one embodiment of the present invention, when the updated vector velocity of each particle is less than the vector velocity threshold, it is determined that the updated vector velocity of the particle satisfies the vector velocity threshold.

[0127] Step 2058: When it is determined that the vector velocity threshold is met, the global optimal solution is determined as the optimal solution.

[0128] In one embodiment of the present invention, when the updated vector data of all particles are less than the vector velocity threshold, it is determined that the current particle swarm algorithm has converged. Therefore, the current global optimal solution is determined as the optimal solution in terms of patience.

[0129] Step 30: Determine the target user's predicted interest in each of the predicted attractions based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each attraction to be predicted.

[0130] In one embodiment of the present invention, in order to apply the marginal effect, in addition to the patience of the target user, it is also necessary to determine the theoretical number of visits to the attraction to be predicted. Then, the initial interest in the attraction to be predicted is weighted based on the theoretical number of visits and the patience, to obtain the predicted interest. The initial interest in the attraction to be predicted can be determined based on its attraction feature information, and the theoretical number of visits can be determined based on the similarity between the attraction to be predicted and historical tourist attractions. The higher the similarity, the greater the probability that the target user will visit the attraction to be predicted based on their previous visits to historical tourist attractions.

[0131] Therefore, in another embodiment of the present invention, step 30 further includes:

[0132] Step 301: Determine the similarity between the historical tourist attraction and the attraction to be predicted based on the attraction feature information of the historical tourist attraction and the attraction feature information of the attraction to be predicted.

[0133] In one embodiment of the present invention, the distance between the feature information of historical tourist attractions and the tourist attraction to be predicted is calculated, and the similarity is determined based on the reciprocal of the distance.

[0134] Step 302: Determine the predicted number of visits to the tourist attraction based on the similarity.

[0135] In one embodiment of the present invention, the similarity between the predicted tourist attraction and the historical tourist attraction can be regarded as a vote on the number of times the tourist attraction to be predicted will be visited. That is, the higher the similarity between the tourist attraction to be predicted and the user's historical tourist attractions, the greater the likelihood that the user will visit the tourist attraction to be predicted. Therefore, the predicted number of times the tourist attraction to be predicted can be determined based on the sum of the similarities between all historical tourist attractions.

[0136] For example, if the historical tourist attraction is Qingdao, and the predicted attraction is Yantai, both Qingdao and Yantai are known for their seaside and seafood tourism. Therefore, people who have visited Qingdao in the past are more likely to choose to visit Yantai in the future than those who have not visited Qingdao. This means that consumption behavior between different consumer goods can also influence each other, and this influence is determined by the similarity between the goods.

[0137] Therefore, in another embodiment of the present invention, step 302 further includes:

[0138] Step 3021: Determine the sum of the similarities between the predicted attraction and all the historical tourist attractions.

[0139] Step 3022: Determine the predicted number of trips based on the sum of the similarities.

[0140] In one embodiment of the invention, the sum of similarities is determined as the predicted number of trips.

[0141] Step 303: Determine the initial interest level of the attraction to be predicted.

[0142] In one embodiment of the present invention, the initial interest level of the attraction to be predicted can be determined by referring to the collaborative filtering method in step 203, which will not be described in detail here.

[0143] Step 304: Determine the predicted interest level of each of the attractions to be predicted based on the patience level, the predicted number of trips, and the initial interest level of the attractions to be predicted.

[0144] In one embodiment of the present invention, step 304 may include:

[0145] The predicted interest level of the scenic spot to be predicted is determined according to the following formula:

[0146]

[0147] Where E is the predicted interest level of the attraction to be predicted, E' is the initial interest level of the attraction to be predicted, t is the predicted number of trips, and σ is the patience level.

[0148] Similar to the formula used in step 2041 to calculate the actual interest level of historical tourist attractions, the predicted interest level of the attraction to be predicted can be obtained by weighting it using a Gaussian function based on its initial interest level, the predicted number of visits, and patience level. The difference is that the predicted number of visits for the attraction to be predicted is determined based on its similarity to all historical tourist attractions, while in step 2041, the historical number of visits for historical tourist attractions is determined based on the tourism time information of the attraction category it belongs to.

[0149] Step 40: Determine the target recommended attraction for the target user based on the predicted interest level of each of the attractions to be predicted.

[0150] In one embodiment of the present invention, tourist attractions with a predicted interest level higher than a certain threshold can be identified as target recommended tourist attractions. Furthermore, the tourist attractions can be sorted from high to low according to their predicted interest level to obtain a tourist attraction recommendation sequence and output it to the target user.

[0151] In another embodiment of the present invention, the attractions to be predicted and historical tourist attractions can be comprehensively sorted, and the predicted interest and actual interest values ​​can be sorted from high to low to obtain a recommended sequence of attractions and output to the target user.

[0152] The information recommendation method provided in this invention determines the target user's historical travel trajectory data, which corresponds to multiple historical tourist attractions. Then, it determines the target user's patience level based on the historical travel trajectory data. The patience level characterizes the degree to which the number of times a target user visits a historical tourist attraction affects their interest in that attraction; that is, the lower the patience level, the greater the decrease in the target user's interest in that historical tourist attraction as the number of visits increases. Based on the patience level, the attraction's characteristic information, and the characteristic information of each attraction to be predicted, the target user's predicted interest level for each attraction to be predicted is determined. Finally, based on the predicted interest level of each attraction to be predicted, the target recommended attraction for the target user is determined.

[0153] This invention differs from existing technologies that use collaborative filtering, which results in low accuracy in recommending tourist attractions. Instead, it mines data from users' historical travel patterns to determine the impact of the number of times a user visits a historical tourist attraction on the target user's interest in that attraction—that is, the target user's patience. This patience is then combined with the results of collaborative filtering to obtain a comprehensive predicted interest level, thereby improving the accuracy of tourist attraction recommendations.

[0154] Figure 2 A schematic diagram of the information recommendation device provided in an embodiment of the present invention is shown. Figure 3 As shown, the device 500 includes: a first determination module 501, a second determination module 502, a prediction module 503, and a recommendation module 504.

[0155] The first determining module 501 is used to determine the historical travel trajectory data of the target user; the historical travel trajectory data corresponds to multiple historical tourist attractions.

[0156] The second determining module 502 is used to determine the patience level of the target user based on the historical travel trajectory data; the patience level is used to characterize the degree of influence of the number of times the target user visits the historical tourist attractions on the target user's interest in the historical tourist attractions;

[0157] The prediction module 503 is used to determine the target user's predicted interest in each of the predicted attractions based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each attraction to be predicted.

[0158] The recommendation module 504 is used to determine the target recommended attractions corresponding to the target user based on the predicted interest level of each of the attractions to be predicted.

[0159] The operation process of the information recommendation device in this embodiment of the invention is largely the same as that of the aforementioned information recommendation method embodiment, and will not be described again.

[0160] The information recommendation device provided in this embodiment of the invention determines the historical travel trajectory data of a target user; the historical travel trajectory data corresponds to multiple historical tourist attractions; then, it determines the target user's patience level based on the historical travel trajectory data; wherein, the patience level is used to characterize the degree to which the number of times the target user visits a historical tourist attraction affects the target user's interest in that historical tourist attraction, that is, the lower the patience level, the greater the decrease in the target user's interest in that historical tourist attraction as the number of times the target user visits it increases; the predicted interest level of the target user in each of the predicted attractions is determined based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each attraction to be predicted; and the target recommended attractions corresponding to the target user are determined based on the predicted interest level of each attraction to be predicted.

[0161] This invention distinguishes itself from existing technologies that use collaborative filtering, which results in low accuracy in recommending tourist attractions. The information recommendation device provided in this embodiment of the invention mines the influence of the number of times a user has visited a historical tourist attraction on the target user's interest in that attraction from the user's historical travel trajectory data, i.e., the target user's patience. This patience is then combined with the results of collaborative filtering to obtain the predicted interest, thereby improving the accuracy of tourist attraction recommendations.

[0162] Figure 3 The diagram shows a structural schematic of an information recommendation device provided in an embodiment of the present invention. The specific embodiments of the present invention do not limit the specific implementation of the information recommendation device.

[0163] like Figure 3As shown, the information recommendation device may include: a processor 602, a communications interface 604, a memory 606, and a communications bus 608.

[0164] The processor 602, communication interface 604, and memory 606 communicate with each other via communication bus 608. Communication interface 604 is used to communicate with other network elements such as clients or other servers. The processor 602 executes program 610, specifically performing the relevant steps described above in the information recommendation method embodiment.

[0165] Specifically, program 610 may include program code, which includes computer-executable instructions.

[0166] Processor 602 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The information recommendation device may include one or more processors of the same type, such as one or more CPUs; or it may include processors of different types, such as one or more CPUs and one or more ASICs.

[0167] Memory 606 is used to store program 610. Memory 606 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0168] Specifically, program 610 can be called by processor 602 to cause the information recommendation device to perform the following operations:

[0169] Determine the target user's historical travel trajectory data; the historical travel trajectory data corresponds to multiple historical tourist attractions;

[0170] The patience level of the target user is determined based on the historical travel trajectory data; the patience level is used to characterize the degree to which the number of times the target user visits the historical tourist attractions affects the target user's interest in the historical tourist attractions.

[0171] The target user's predicted interest in each of the predicted attractions is determined based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each attraction to be predicted.

[0172] The target recommended attractions for the target user are determined based on the predicted interest level of each of the attractions to be predicted.

[0173] The operation process of the information recommendation device in this embodiment of the invention is largely the same as that of the aforementioned information recommendation method embodiment, and will not be described again.

[0174] The information recommendation device provided in this embodiment of the invention determines the historical travel trajectory data of a target user; the historical travel trajectory data corresponds to multiple historical tourist attractions; then, it determines the target user's patience level based on the historical travel trajectory data; wherein, the patience level is used to characterize the degree to which the target user's number of visits to a historical tourist attraction affects the target user's interest in that historical tourist attraction, that is, the lower the patience level, the greater the decrease in the target user's interest in that historical tourist attraction as the number of visits to that historical tourist attraction increases; the predicted interest level of the target user in each of the predicted attractions is determined based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each attraction to be predicted; and the target recommended attraction corresponding to the target user is determined based on the predicted interest level of each attraction to be predicted.

[0175] This invention distinguishes itself from existing technologies that use collaborative filtering, which results in low accuracy in recommending tourist attractions. The information recommendation device provided in this embodiment of the invention mines the influence of the number of times a user has visited a historical tourist attraction on the target user's interest in that attraction from the user's historical travel trajectory data, i.e., the target user's patience. This patience is then combined with the results of collaborative filtering to obtain the predicted interest, thereby improving the accuracy of tourist attraction recommendations.

[0176] This invention provides a computer-readable storage medium storing at least one executable instruction that, when executed on an information recommendation device, causes the information recommendation device to perform the information recommendation method described in any of the above method embodiments.

[0177] Specifically, the executable instructions can be used to cause the information recommendation device to perform the following operations:

[0178] Determine the target user's historical travel trajectory data; the historical travel trajectory data corresponds to multiple historical tourist attractions;

[0179] The patience level of the target user is determined based on the historical travel trajectory data; the patience level is used to characterize the degree to which the number of times the target user visits the historical tourist attractions affects the target user's interest in the historical tourist attractions.

[0180] The target user's predicted interest in each of the predicted attractions is determined based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each attraction to be predicted.

[0181] The target recommended attractions for the target user are determined based on the predicted interest level of each of the attractions to be predicted.

[0182] The operation process of the computer-readable storage medium in this embodiment of the invention is largely the same as that in the aforementioned information recommendation method embodiment, and will not be described again.

[0183] The computer storage medium provided in this embodiment of the invention determines the historical travel trajectory data of a target user; the historical travel trajectory data corresponds to multiple historical tourist attractions; then, it determines the target user's patience level based on the historical travel trajectory data; wherein, the patience level is used to characterize the degree to which the number of times the target user visits a historical tourist attraction affects the target user's interest in that historical tourist attraction, that is, the lower the patience level, the greater the decrease in the target user's interest in that historical tourist attraction as the number of times the target user visits it increases; the predicted interest level of the target user in each of the predicted attractions is determined based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each attraction to be predicted; and the target recommended attractions corresponding to the target user are determined based on the predicted interest level of each attraction to be predicted.

[0184] This invention distinguishes itself from existing technologies that use collaborative filtering, which results in low accuracy in recommending tourist attractions. The information recommendation device provided in this embodiment of the invention mines the influence of the number of times a user has visited a historical tourist attraction on the target user's interest in that attraction from the user's historical travel trajectory data, i.e., the target user's patience. This patience is then combined with the results of collaborative filtering to obtain the predicted interest, thereby improving the accuracy of tourist attraction recommendations.

[0185] This invention provides an information recommendation device for performing the above-described information recommendation method.

[0186] This invention provides a computer program that can be called by a processor to cause an information recommendation device to execute the information recommendation method in any of the above method embodiments.

[0187] This invention provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions that, when executed on a computer, cause the computer to perform the information recommendation method described in any of the above method embodiments.

[0188] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, the embodiments of the present invention are not directed to any particular programming language. It should be understood that the content of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0189] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0190] Similarly, it should be understood that, in order to simplify the invention and aid in understanding one or more of the various aspects of the invention, features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the above description of exemplary embodiments of the invention. However, this disclosure should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in the various claims.

[0191] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, various features disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by alternative features that serve the same, equivalent, or similar purpose.

[0192] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. An information recommendation method, characterized in that, The method includes: Determine the target user's historical travel trajectory data; the historical travel trajectory data corresponds to multiple historical tourist attractions; wherein, the historical travel trajectory data includes the attraction feature information and travel time information of each of the historical tourist attractions; Based on the scenic spot characteristics information, the historical tourist attractions are divided into at least one scenic spot category; each scenic spot category includes at least one of the historical tourist attractions; the number of times the target user has visited each scenic spot category is determined based on the travel time information; the initial interest level of each historical tourist attraction is determined; the total interest level of the historical travel trajectory data is determined based on the initial interest level of each historical tourist attraction, the number of times it has visited its scenic spot category, and the initial patience level of the target user; the initial patience level is optimized with maximizing the total interest level as the optimization objective, and the optimal solution is determined as the patience level; the patience level is used to characterize the degree of influence of the number of times the target user has visited the historical tourist attractions on the target user's interest level in the historical tourist attractions. The determination of the total interest level of the historical travel trajectory data based on the initial interest level of each historical tourist attraction, the number of historical visits to the attraction category, and the initial patience level of the target user further includes: The actual level of interest in each of the aforementioned historical tourist attractions is determined according to the following formula; ； in, Any one of the historical tourist attractions mentioned above. express The first of all the historical tourist attractions mentioned The target users are traveling. for The actual level of interest, for The initial degree of interest, for The number of historical visits to the attraction category in question. The initial patience level; The total interest score is determined based on the actual interest scores of all the historical tourist attractions in the historical tourism trajectory data. The target user's predicted interest in each of the predicted attractions is determined based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each attraction to be predicted. The target recommended attractions for the target user are determined based on the predicted interest level of each of the attractions to be predicted.

2. The method according to claim 1, characterized in that, The optimization of the initial patience level with the goal of maximizing the total interest value, and the determination of the optimal solution as the patience level, includes: Multiple particles and their corresponding initial velocity vectors are randomly generated, wherein each particle corresponds to a patience value. Based on the patience level corresponding to each particle, the total interest level corresponding to each particle is determined. The total interest is determined as a reference value for each of the particles; Based on the reference values, determine the local optimal solution corresponding to each particle and the global optimal solution corresponding to all particles; The initial velocity vectors of each particle are updated based on the local optimal solution and the global optimal solution to obtain the updated velocity vectors of each particle. The updated position of each particle and the patience level corresponding to the updated position are determined based on the current position of each particle and the updated velocity vector. Determine whether the updated vector velocity of each particle meets the vector velocity threshold; When the vector velocity threshold is satisfied, the global optimal solution is determined as the optimal solution.

3. The method according to claim 1, characterized in that, The step of determining the target user's predicted interest level in each of the predicted attractions based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each attraction to be predicted includes: The similarity between the historical tourist attraction and the attraction to be predicted is determined based on the attraction feature information of the historical tourist attraction and the attraction to be predicted. The predicted number of visits to the destination is determined based on the similarity. Determine the initial level of interest in the tourist attraction to be predicted; The predicted interest level of each of the predicted attractions is determined based on the patience level, the predicted number of trips, and the initial interest level of the attractions to be predicted.

4. The method according to claim 3, characterized in that, Determining the predicted number of visits to the destination based on the similarity includes: Determine the sum of the similarities between the proposed attraction and all the historical tourist attractions; The predicted number of trips is determined based on the sum of the similarities.

5. The method according to claim 3, characterized in that, The step of determining the predicted interest level of each of the predicted attractions based on the patience level, the predicted number of trips, and the initial interest level of each of the predicted attractions includes: The predicted interest level of the scenic spot to be predicted is determined according to the following formula: ； in, The predicted interest level of the scenic spot to be predicted. The initial interest level of the scenic spot to be predicted is... For the predicted number of trips, The patience level mentioned above.

6. An information recommendation device, characterized in that, The device includes: The first determining module is used to determine the historical travel trajectory data of the target user; the historical travel trajectory data corresponds to multiple historical tourist attractions; wherein, the historical travel trajectory data includes the attraction feature information and travel time information of each of the historical tourist attractions; The second determining module is used to classify the historical tourist attractions into at least one attraction class based on the attraction feature information of the historical tourist attractions; the attraction class includes at least one of the historical tourist attractions; determine the number of times the target user has visited each of the attraction classes based on the travel time information; determine the initial interest level of each of the historical tourist attractions; determine the total interest level of the historical travel trajectory data based on the initial interest level of each of the historical tourist attractions, the number of times the attraction class belongs to, and the initial patience level of the target user; optimize the initial patience level with the maximum total interest level as the optimization objective, and determine the optimal solution as the patience level; the patience level is used to characterize the degree of influence of the number of times the target user visits the historical tourist attractions on the target user's interest level of the historical tourist attractions; wherein, determining the total interest level of the historical travel trajectory data based on the initial interest level of each of the historical tourist attractions, the number of times the attraction class belongs to, and the initial patience level of the target user further includes: determining the actual interest level of each of the historical tourist attractions based on the following formula. ； in, Any one of the historical tourist attractions mentioned above. express The first of all the historical tourist attractions mentioned The target users are traveling. for The actual level of interest, for The initial degree of interest, for The number of historical visits to the attraction category in question. The initial patience level is defined as follows: the total interest level is determined based on the actual interest level values ​​of all the historical tourist attractions in the historical travel trajectory data; the prediction module is used to determine the predicted interest level of the target user for each of the tourist attractions to be predicted based on the patience level, the attraction feature information of the historical tourist attractions, and the attraction feature information of each tourist attraction to be predicted. The recommendation module is used to determine the target recommended attractions corresponding to the target user based on the predicted interest level of each of the attractions to be predicted.

7. An information recommendation device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction that causes the processor to perform the operation of the information recommendation method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The storage medium stores at least one executable instruction, which, when executed on the information recommendation device, causes the information recommendation device to perform the operation of the information recommendation method as described in any one of claims 1-5.

Images