Method and apparatus for predicting user preference, electronic device, and storage medium

Abstract

Embodiments of the present application provide a user preference prediction method and device, electronic equipment and storage medium, relating to the technical field of data processing, the method comprising: obtaining first evaluation information of a target user on a target equity product purchased historically; determining a target commodity of the same category as the target equity product and obtaining second evaluation information of the target user on the target commodity; calculating a first evaluation sentiment value corresponding to the first evaluation information and a second evaluation sentiment value corresponding to the second evaluation information, and performing feature integration on the first evaluation sentiment value and the second evaluation sentiment value to obtain a first emotion feature vector corresponding to the target equity product and a second emotion feature vector corresponding to the target commodity; obtaining a user equity parameter of the target user, and performing product prediction according to the first emotion feature vector, the second emotion feature vector and the user equity parameter to obtain the target equity product, thereby effectively recommending an equity product that is more in line with the user's preference.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method for predicting user preferences, a device for predicting user preferences, an electronic device, and a computer-readable storage medium. Background Technology

[0002] Information push is a technology that uses certain technical standards or protocols to reduce information overload by pushing relevant information to users on the internet. By proactively sending information to users, information push technology can reduce the time users spend searching online. In existing technologies, operators typically push information about user benefits to facilitate access to better communication services. However, due to inaccurate targeting of the content or users, some users receive push information that is not what they need, causing inconvenience and resulting in low effectiveness of the pushed information. Summary of the Invention

[0003] The present invention provides a method, apparatus, electronic device, and computer-readable storage medium for predicting user preferences, in order to solve or partially solve the problem that the effectiveness of information push is poor because the rights and benefits recommendation cannot meet user needs.

[0004] This invention discloses a method for predicting user preferences, comprising:

[0005] Obtain the target users' initial evaluation information on the target benefit products they have purchased in the past;

[0006] Identify target products that belong to the same category as the target benefit product, and obtain the second evaluation information of the target user on the target product;

[0007] Calculate the first evaluation sentiment value corresponding to the first evaluation information and the second evaluation sentiment value corresponding to the second evaluation information, and integrate the first evaluation sentiment value and the second evaluation sentiment value to obtain a first sentiment feature vector corresponding to the target benefit product and a second sentiment feature vector corresponding to the target product;

[0008] Obtain the user rights parameters of the target user, and perform product prediction based on the first sentiment feature vector, the second sentiment feature vector, and the user rights parameters to obtain the target rights product.

[0009] In some feasible implementations, calculating the first evaluation sentiment value corresponding to the first evaluation information includes:

[0010] The first evaluation information is segmented into words to obtain the first segmentation information;

[0011] Part-of-speech tagging is performed on the first segmented word information to obtain the part-of-speech tagging result of the first segmented word information;

[0012] Using the first word segmentation information and the part-of-speech tagging result, word groups are extracted from the first evaluation information to obtain the first word group set corresponding to the first evaluation information;

[0013] Obtain the number of times the first word group set appears in the historical evaluation information published by the target user, and calculate the sentiment value of the first word group corresponding to each first word group set;

[0014] The first evaluation sentiment value of the first evaluation information is obtained by calculating the sentiment values ​​of each of the first word groups.

[0015] In some feasible implementations, the historical evaluation information includes at least positive and negative evaluation information. The step of obtaining the frequency of occurrence of the first phrase set in the historical evaluation information posted by the target user, and calculating the sentiment value of the first phrase corresponding to each first phrase set, includes:

[0016] Obtain the first occurrence count of the first phrase set in the positive review information;

[0017] Obtain the second occurrence count of the first phrase set in the negative review information;

[0018] The sentiment value of the first phrase corresponding to each set of the first phrases is obtained by calculating the first occurrence count and the second occurrence count using the following formula:

[0019]

[0020] Wherein, N1 is the first occurrence count, and N2 is the second occurrence count; if either N1 or N2 is 0, then the sentiment value of the corresponding first word set is not calculated.

[0021] In some feasible implementations, calculating the second evaluation sentiment value corresponding to the second evaluation information includes:

[0022] The second evaluation information is segmented into words to obtain the second segmentation information;

[0023] Part-of-speech tagging is performed on the second word segmentation information to obtain the part-of-speech tagging result of the second word segmentation information;

[0024] Using the second word segmentation information and the part-of-speech tagging result, word groups are extracted from the second evaluation information to obtain the second word group set corresponding to the second evaluation information;

[0025] Obtain the number of times the second word set appears in the historical evaluation information published by the target user, and calculate the sentiment value of the second word set corresponding to each second word set;

[0026] The second evaluation sentiment value of the second evaluation information is obtained by calculating the sentiment values ​​of each of the second phrases.

[0027] In some feasible implementations, the historical evaluation information includes at least positive and negative evaluation information. The step of obtaining the frequency of occurrence of the second word set in the historical evaluation information posted by the target user, and calculating the sentiment value of the second word set corresponding to each second word set, includes:

[0028] Obtain the third occurrence count of the second phrase set in the positive review information;

[0029] Obtain the fourth occurrence count of the second phrase set in the negative review information;

[0030] The sentiment value of the second phrase corresponding to each set of the second phrase is obtained by calculating the third occurrence count and the fourth occurrence count using the following formula:

[0031]

[0032] Wherein, N1 is the third occurrence count, and N2 is the fourth occurrence count; if either N1 or N2 is 0, then the sentiment value of the corresponding second word set is not calculated.

[0033] In some feasible implementations, the step of integrating the first evaluation sentiment value and the second evaluation sentiment value to obtain a first sentiment feature vector corresponding to the target benefit product and a second sentiment feature vector corresponding to the target product includes:

[0034] The average of the first evaluation sentiment values ​​of each of the first evaluation information is taken as the sentiment value of the target benefit product, and the average of the second evaluation sentiment values ​​of each of the second evaluation information is taken as the sentiment value of the target product.

[0035] Obtain the equity product characteristics of the target equity product;

[0036] The first emotional feature vector corresponding to the target equity product is obtained by concatenating the equity product features, product category and emotional value of the target equity product.

[0037] The product category and the emotional value of the target product are concatenated to obtain the second emotional feature vector corresponding to the target product.

[0038] In some feasible implementations, the user benefit parameters include at least the number of benefit views, transaction amount, and number of transactions. The step of predicting the target benefit product based on the first sentiment feature vector, the second sentiment feature vector, and the user benefit parameters includes:

[0039] The number of times the rights and interests are viewed, the transaction amount, the number of transactions, the first sentiment feature vector, and the second sentiment feature vector are output into a preset rights and interests prediction model to obtain the target rights and interests product.

[0040] This invention also discloses a user preference prediction device, comprising:

[0041] The product acquisition module is used to obtain the target user's first evaluation information on the target benefit products purchased in the past.

[0042] The evaluation information acquisition module is used to determine the target product that is of the same category as the target benefit product, and to acquire the second evaluation information of the target user on the target product;

[0043] The vector determination module is used to calculate the first evaluation sentiment value corresponding to the first evaluation information and the second evaluation sentiment value corresponding to the second evaluation information, and to integrate the first evaluation sentiment value and the second evaluation sentiment value to obtain a first sentiment feature vector corresponding to the target benefit product and a second sentiment feature vector corresponding to the target product.

[0044] The product prediction module is used to obtain the user rights parameters of the target user, and to perform product prediction based on the first sentiment feature vector, the second sentiment feature vector and the user rights parameters to obtain the target rights product.

[0045] In some feasible implementations, the vector determination module is specifically used for:

[0046] The first evaluation information is segmented into words to obtain the first segmentation information;

[0047] Part-of-speech tagging is performed on the first segmented word information to obtain the part-of-speech tagging result of the first segmented word information;

[0048] Using the first word segmentation information and the part-of-speech tagging result, word groups are extracted from the first evaluation information to obtain the first word group set corresponding to the first evaluation information;

[0049] Obtain the number of times the first word group set appears in the historical evaluation information published by the target user, and calculate the sentiment value of the first word group corresponding to each first word group set;

[0050] The first evaluation sentiment value of the first evaluation information is obtained by calculating the sentiment values ​​of each of the first word groups.

[0051] In some feasible implementations, the historical evaluation information includes at least positive and negative evaluation information, and the vector determination module is specifically used for:

[0052] Obtain the first occurrence count of the first phrase set in the positive review information;

[0053] Obtain the second occurrence count of the first phrase set in the negative review information;

[0054] The sentiment value of the first phrase corresponding to each set of the first phrases is obtained by calculating the first occurrence count and the second occurrence count using the following formula:

[0055]

[0056] Wherein, N1 is the first occurrence count, and N2 is the second occurrence count; if either N1 or N2 is 0, then the sentiment value of the corresponding first word set is not calculated.

[0057] In some feasible implementations, the vector determination module is specifically used for:

[0058] The second evaluation information is segmented into words to obtain the second segmentation information;

[0059] Part-of-speech tagging is performed on the second word segmentation information to obtain the part-of-speech tagging result of the second word segmentation information;

[0060] Using the second word segmentation information and the part-of-speech tagging result, word groups are extracted from the second evaluation information to obtain the second word group set corresponding to the second evaluation information;

[0061] Obtain the number of times the second word set appears in the historical evaluation information published by the target user, and calculate the sentiment value of the second word set corresponding to each second word set;

[0062] The second evaluation sentiment value of the second evaluation information is obtained by calculating the sentiment values ​​of each of the second phrases.

[0063] In some feasible implementations, the historical evaluation information includes at least positive and negative evaluation information, and the vector determination module is specifically used for:

[0064] Obtain the third occurrence count of the second phrase set in the positive review information;

[0065] Obtain the fourth occurrence count of the second phrase set in the negative review information;

[0066] The sentiment value of the second phrase corresponding to each set of the second phrase is obtained by calculating the third occurrence count and the fourth occurrence count using the following formula:

[0067]

[0068] Wherein, N1 is the third occurrence count, and N2 is the fourth occurrence count; if either N1 or N2 is 0, then the sentiment value of the corresponding second word set is not calculated.

[0069] In some feasible implementations, the vector determination module is specifically used for:

[0070] The average of the first evaluation sentiment values ​​of each of the first evaluation information is taken as the sentiment value of the target benefit product, and the average of the second evaluation sentiment values ​​of each of the second evaluation information is taken as the sentiment value of the target product.

[0071] Obtain the equity product characteristics of the target equity product;

[0072] The first emotional feature vector corresponding to the target equity product is obtained by concatenating the equity product features, product category and emotional value of the target equity product.

[0073] The product category and the emotional value of the target product are concatenated to obtain the second emotional feature vector corresponding to the target product.

[0074] In some feasible implementations, the user rights parameters include at least the number of rights browsing sessions, transaction amount, and number of transactions. The product prediction module is specifically used for:

[0075] The number of times the rights and interests are viewed, the transaction amount, the number of transactions, the first sentiment feature vector, and the second sentiment feature vector are output into a preset rights and interests prediction model to obtain the target rights and interests product.

[0076] This invention also discloses an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0077] The memory is used to store computer programs;

[0078] When the processor executes a program stored in the memory, it implements the method described in the embodiments of the present invention.

[0079] This invention also discloses a computer-readable storage medium storing instructions that, when executed by one or more processors, cause the processors to perform the methods described in this invention.

[0080] The embodiments of the present invention have the following advantages:

[0081] In this embodiment of the invention, by obtaining the first evaluation information of the target user on the target benefit product purchased in the past, then determining the target product of the same category as the target benefit product, and obtaining the second evaluation information of the target user on the target product, the first evaluation sentiment value corresponding to the first evaluation information and the second evaluation sentiment value corresponding to the second evaluation information are calculated. The first evaluation sentiment value and the second evaluation sentiment value are then integrated to obtain the first sentiment feature vector corresponding to the target benefit product and the second sentiment feature vector corresponding to the target product. Then, the user benefit parameters of the target user are obtained, and product prediction is performed based on the first sentiment feature vector, the second sentiment feature vector, and the user benefit parameters to obtain the target benefit product. Thus, by determining the target product of the same category as the target benefit product purchased in the past and obtaining the second evaluation information of the target product, data supplementation is achieved, making the calculation of sentiment values ​​more accurate and improving the overall recommendation accuracy. At the same time, product prediction based on sentiment values ​​can effectively recommend benefit products that are more in line with user preferences, which not only meets user needs but also improves the effectiveness of information push. Attached Figure Description

[0082] Figure 1 This is a flowchart illustrating the steps of a user preference prediction method provided in an embodiment of the present invention.

[0083] Figure 2 This is a structural block diagram of a user preference prediction device provided in an embodiment of the present invention;

[0084] Figure 3 This is a block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0085] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0086] As an example, with the development of internet technology, more and more users prefer online shopping. To better recommend relevant content to users, it's common practice to collect user preference information and recommend relevant benefits based on this information, thus facilitating a better communication experience. However, due to inaccurate targeting of content or users, some users receive push notifications that are not what they need, causing inconvenience and resulting in low effectiveness of the pushed information.

[0087] In this invention, during the process of recommending benefit products to users, the process involves obtaining the target user's first evaluation information on historically purchased target benefit products, then identifying target products of the same category as the target benefit product, and obtaining the target user's second evaluation information on the target product. The process then calculates the first evaluation sentiment value corresponding to the first evaluation information and the second evaluation sentiment value corresponding to the second evaluation information. These sentiment values ​​are then integrated to obtain a first sentiment feature vector corresponding to the target benefit product and a second sentiment feature vector corresponding to the target product. Furthermore, the user's benefit parameters are obtained, and product prediction is performed based on the first and second sentiment feature vectors and the user benefit parameters to determine the target benefit product. By identifying target products of the same category as the user's historically purchased benefit products and obtaining the second evaluation information of the target products, data supplementation is achieved, making the calculation of sentiment values ​​more accurate and improving the overall recommendation accuracy. Simultaneously, product prediction based on sentiment values ​​can effectively recommend benefit products that better match user preferences, satisfying user needs and improving the effectiveness of information push.

[0088] It should be noted that the recommendation process may involve the collection of user data, which can be carried out in accordance with local laws and with the user's authorization to ensure the security of the user data.

[0089] Reference Figure 1 The diagram illustrates a flowchart of a user preference prediction method provided in an embodiment of the present invention, which may specifically include the following steps:

[0090] Step 101: Obtain the target user's first evaluation information on the target benefit products purchased in the past;

[0091] Optionally, the present invention can be applied to applications that can provide users with benefits and products. During operation, the application can acquire relevant user data and recommend corresponding benefits and products to users based on the acquired data, so that users can enjoy suitable services.

[0092] In its implementation, the application can obtain the target benefit products that the user has previously purchased, as well as the user's initial evaluation information for those target benefit products. By obtaining the target benefit products, it can determine which categories of benefit products the user prefers, and by obtaining the initial evaluation information, it can determine the user's preference for the benefit products they have purchased. Specifically, it can obtain benefit product data purchased by the user within a preset time period (e.g., within one week, one month, three months, one year, etc.) and extract the user's initial evaluation information for each benefit product from the benefit product data. The user can have multiple initial evaluations for each benefit product. For example, if the user purchased benefit products A, B, and C in the past year, the user might have two evaluations for benefit product A (A1 and A2), one evaluation for benefit product B (B1), and three evaluations for benefit product C (C1, C2, and C3).

[0093] Step 102: Determine the target product as being in the same category as the target benefit product, and obtain the second evaluation information of the target user on the target product;

[0094] Since the number of benefit products a single user has purchased historically is limited, and therefore the amount of review information is also limited, the data volume is insufficient for accurate recommendations. Therefore, in this embodiment of the invention, data is supplemented by obtaining user reviews of other products in the same category as the purchased benefit products. Specifically, the second set of review information can be user reviews of target products in the same category as the target benefit product.

[0095] In practice, the first step is to determine the category of the primary benefit product, such as music, games, education, video, fitness, social networking, shopping, etc. Then, the target products that the user has already purchased and that belong to the same category as the target benefit product are identified. Finally, the user's evaluation information of the target products is obtained as the second evaluation information.

[0096] For example, if the benefit product A purchased by a user is a video product, then the information on the video products purchased by the user, such as projectors, screens, audio equipment, movie tickets, and DVDs, is obtained, and the user's evaluation information on the above products is obtained, which is the second evaluation information.

[0097] Step 103: Calculate the first evaluation sentiment value corresponding to the first evaluation information and the second evaluation sentiment value corresponding to the second evaluation information, and integrate the first evaluation sentiment value and the second evaluation sentiment value to obtain the first sentiment feature vector corresponding to the target benefit product and the second sentiment feature vector corresponding to the target product.

[0098] Once the first evaluation information of the target benefit product and the second evaluation information of the target goods of the same category as the target benefit product are determined, the first evaluation sentiment value corresponding to the first evaluation information and the second evaluation sentiment value corresponding to the second evaluation information can be calculated respectively. The first and second evaluation sentiment values ​​are then integrated to obtain a first sentiment feature vector corresponding to the target benefit product and a second sentiment feature vector corresponding to the target goods. This allows for product prediction based on the obtained sentiment feature vectors. The first evaluation sentiment value can represent the sentiment tendency of the first evaluation information, and the second evaluation sentiment value can represent the sentiment tendency of the second evaluation information, such as a positive, neutral, or negative tendency, etc. This invention does not impose any limitations on this.

[0099] In some feasible implementations, both the first and second evaluation information can be text information. The first evaluation information can be segmented into words to obtain first segmentation information. Then, part-of-speech tagging (POS) can be performed on the first segmentation information to obtain the POS tagging results. Using the first segmentation information and the POS tagging results, phrases can be extracted from the first evaluation information to obtain the first phrase set corresponding to the first evaluation information. The frequency of occurrence of the first phrase set in the target user's historical evaluation information can be obtained. The sentiment value of the first phrase set corresponding to each first phrase set can be calculated. Then, the sentiment value of the first evaluation information can be obtained by using the sentiment values ​​of each phrase set. The historical evaluation information includes at least positive and negative evaluation information. Therefore, the first occurrence frequency of the first phrase set in positive evaluation information and the second occurrence frequency in negative evaluation information can be obtained. Then, the sentiment value of the first phrase set corresponding to each first phrase set can be obtained by using the first and second occurrence frequencies according to the following formula:

[0100]

[0101] Where N1 is the first occurrence count and N2 is the second occurrence count; if either N1 or N2 is 0, the sentiment value of the corresponding first word set is not calculated.

[0102] Correspondingly, for the second evaluation information, we can first segment the second evaluation information to obtain second segmented information, and then perform part-of-speech tagging on the second segmented information to obtain the part-of-speech tagging results. Next, using the second segmented information and the part-of-speech tagging results, we can extract phrases from the second evaluation information to obtain the second phrase set corresponding to the second evaluation information. Then, we can obtain the number of times the second phrase set appears in the target user's historical evaluation information, calculate the phrase sentiment value corresponding to each second phrase set, and then use the phrase sentiment values ​​to calculate the second evaluation sentiment value of the second evaluation information. Since the historical evaluation information includes at least positive and negative evaluation information, we can obtain the third occurrence count of the second phrase set in positive evaluation information and the fourth occurrence count of the second phrase set in negative evaluation information. We can then use the third and fourth occurrence counts to calculate the phrase sentiment value corresponding to each second phrase set according to the following formula:

[0103]

[0104] Where N1 is the third occurrence count and N2 is the fourth occurrence count; if either N1 or N2 is 0, the sentiment value of the corresponding second word set will not be calculated.

[0105] It should be noted that the sentiment value of a phrase can represent the sentiment tendency of a phrase set, such as a positive tendency, a neutral tendency, or a negative tendency, etc., and this invention does not limit this.

[0106] After calculating the first sentiment value corresponding to each first evaluation information and the second sentiment value corresponding to each second evaluation information, the average of the first sentiment values ​​of each first evaluation information can be used as the sentiment value of the target benefit product, and the average of the second sentiment values ​​of each second evaluation information can be used as the sentiment value of the target product. At the same time, the benefit product characteristics of the target benefit product are obtained. Then, the benefit product characteristics, product category, and benefit product sentiment value of the target benefit product are concatenated to obtain the first sentiment feature vector corresponding to the target benefit product. Similarly, the product category and sentiment value of the target product are concatenated to obtain the second sentiment feature vector corresponding to the target product. By identifying target products of the same category as the target benefit products purchased by the user in the past and obtaining the second evaluation information of the target products, data supplementation is achieved, making the calculation of sentiment values ​​more accurate and improving the overall recommendation accuracy.

[0107] In one example, both the first and second evaluation information are textual; therefore, the method for calculating the sentiment value is the same. After calculating the sentiment values ​​separately, a sentiment feature vector is formed. The specific process can be as follows:

[0108] S3-1. Perform word segmentation and part-of-speech tagging on the textual information of the first and second evaluation information;

[0109] First, the text information of the evaluation is segmented into words; then, the segmented results are tagged with parts of speech, marking nouns, verbs, adverbs, adjectives, etc. in the text; at the same time, words with positive, negative, or derogatory meanings are also marked.

[0110] For example, a review might say, "The projector has high lumens and the projection is clear, but the built-in speakers are only so-so." The nouns are: projector, lumens, and speakers; the adjectives are: high (positive), clear (positive), and so-so (negative).

[0111] S3-2. Based on the word segmentation results and part-of-speech tagging results, extract word groups from each first evaluation information and second evaluation information text to obtain a word group set.

[0112] Following the rules of "adjective + noun", "adverb + verb", and "noun + adjective", extract phrases from the evaluation text. The extracted phrases must contain words with positive, negative, or derogatory meanings, and then form a set of phrases.

[0113] For example, the extracted phrases were "high lumens", "clear projection", and "average speakers".

[0114] Based on this step, each phrase in the resulting phrase set is a phrase that can reflect the sentiment bias of user evaluation. Subsequently, sentiment values ​​are calculated for the obtained phrases. Phrasees without sentiment bias are not included in the subsequent calculation. This reduces the amount of computation and effectively improves the accuracy of sentiment value calculation.

[0115] S3-3. Calculate the sentiment value of each phrase based on the number of times each phrase appears;

[0116] Calculate the frequency of each phrase in positive and negative user reviews, specifically using a star rating system: five stars and four stars for positive reviews, three stars for neutral reviews, and two stars and one star for negative reviews. The favorability rating (S) is calculated as follows:

[0117]

[0118] Where N1 is the number of times the phrase appears in positive user reviews, and N2 is the number of times the phrase appears in negative user reviews. If either N1 or N2 is 0, the sentiment value of that phrase is not calculated.

[0119] S3-4. Calculate the sentiment value of each first evaluation information or second evaluation information based on the sentiment value of each word group in each first evaluation information or second evaluation information.

[0120] After calculating the sentiment value of each word group, the sentiment values ​​of all word groups in each evaluation are summed to obtain the overall sentiment value of the evaluation.

[0121]

[0122] If the calculated sentiment value Sc for the entire review is greater than 0, then the user's review is positive; otherwise, it is negative; and if it is exactly 0, the review is neutral.

[0123] S3-5. Using rights or product categories as statistical units, calculate the average value of the evaluation sentiment value as the target sentiment value;

[0124] The average sentiment value is calculated based on the statistical evaluation of rights or product categories. For example, for the first evaluation information regarding rights, if a user has two evaluations of product A, the average sentiment value of the two evaluations is calculated as the target sentiment value for product A.

[0125]

[0126] Where n is the number of user reviews.

[0127] As for product categories, there are many products and reviews involved. Similarly, the average sentiment value is calculated. For example, when a user buys video products, the sentiment values ​​of the reviews for products such as projectors, screens, audio equipment, movie tickets, and DVDs are summed and averaged to obtain the user's target sentiment value for video products.

[0128] S3-6. Construct the first equity product-emotion feature vector based on the equity characteristics, equity categories, and corresponding target sentiment values;

[0129] The product features are vu, the category of rights is vc, and the corresponding target sentiment value is Avg(Sc1). The vectors are connected and the dimensions are transformed through a fully connected layer.

[0130]

[0131] Where Wc1 is the weight of the fully connected layer and bc1 is the bias of the fully connected layer.

[0132] S3-7. Based on the product category and the corresponding target sentiment value, construct a second similar product-sentiment feature vector.

[0133] The product category feature is vm, and the corresponding target sentiment value is Avg(Sc2). The vectors are concatenated and the dimensions are transformed through a fully connected layer.

[0134]

[0135] Where Wc2 is the weight of the fully connected layer and bc2 is the bias of the fully connected layer.

[0136] By identifying target products in the same category as the target benefit products purchased by users in their history, and obtaining secondary evaluation information for these target products, data supplementation was achieved, making the calculation of sentiment values ​​more accurate and improving the overall accuracy of recommendations.

[0137] Step 104: Obtain the user rights parameters of the target user, and perform product prediction based on the first emotional feature vector, the second emotional feature vector, and the user rights parameters to obtain the target rights product.

[0138] In specific implementation, user benefit parameters include at least the number of benefit views, transaction amount, and number of transactions. The number of benefit views, transaction amount, number of transactions, the first sentiment feature vector, and the second sentiment feature vector can be used to output a preset benefit prediction model to obtain the target benefit product. By identifying target products of the same category as the user's historically purchased target benefit products and obtaining the second evaluation information of the target products, data supplementation is achieved, making the calculation of sentiment values ​​more accurate and improving the overall recommendation accuracy. Furthermore, product prediction based on sentiment values ​​can effectively recommend benefit products that better match user preferences, satisfying user needs and improving the effectiveness of information push.

[0139] In one example, assuming a user benefit-related feature Ci, including the number of benefit views, transaction amount, and number of transactions, the first feature vector C1, the second feature vector C2, and the user benefit-related feature Ci are input into the benefit prediction model, which outputs the probability of the user's preference for benefit products. Then, information such as the top N benefit content, benefit fees, and benefit discounts that match the user's benefit product preferences is obtained and pushed to the user in real time.

[0140] It should be noted that the embodiments of the present invention include, but are not limited to, the examples described above. It is understood that those skilled in the art can make further settings according to actual needs under the guidance of the ideas in the embodiments of the present invention, and the present invention does not limit such settings.

[0141] In this embodiment of the invention, by obtaining the first evaluation information of the target user on the target benefit product purchased in the past, then determining the target product of the same category as the target benefit product, and obtaining the second evaluation information of the target user on the target product, the first evaluation sentiment value corresponding to the first evaluation information and the second evaluation sentiment value corresponding to the second evaluation information are calculated. The first evaluation sentiment value and the second evaluation sentiment value are then integrated to obtain the first sentiment feature vector corresponding to the target benefit product and the second sentiment feature vector corresponding to the target product. Then, the user benefit parameters of the target user are obtained, and product prediction is performed based on the first sentiment feature vector, the second sentiment feature vector, and the user benefit parameters to obtain the target benefit product. Thus, by determining the target product of the same category as the target benefit product purchased in the past and obtaining the second evaluation information of the target product, data supplementation is achieved, making the calculation of sentiment values ​​more accurate and improving the overall recommendation accuracy. At the same time, product prediction based on sentiment values ​​can effectively recommend benefit products that are more in line with user preferences, which not only meets user needs but also improves the effectiveness of information push.

[0142] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0143] Reference Figure 2 The diagram illustrates a structural block diagram of a user preference prediction device provided in an embodiment of the present invention, which may specifically include the following modules:

[0144] Product acquisition module 201 is used to acquire the target user's first evaluation information on the target benefit products purchased in the past;

[0145] The evaluation information acquisition module 202 is used to determine that the target product is of the same category as the target benefit product, and to acquire the second evaluation information of the target user on the target product;

[0146] The vector determination module 203 is used to calculate the first evaluation sentiment value corresponding to the first evaluation information and the second evaluation sentiment value corresponding to the second evaluation information, and to integrate the first evaluation sentiment value and the second evaluation sentiment value to obtain a first sentiment feature vector corresponding to the target benefit product and a second sentiment feature vector corresponding to the target product.

[0147] The product prediction module 204 is used to obtain the user rights parameters of the target user, and perform product prediction based on the first emotional feature vector, the second emotional feature vector and the user rights parameters to obtain the target rights product.

[0148] In some feasible implementations, the vector determination module 203 is specifically used for:

[0149] The first evaluation information is segmented into words to obtain the first segmentation information;

[0150] Part-of-speech tagging is performed on the first segmented word information to obtain the part-of-speech tagging result of the first segmented word information;

[0151] Using the first word segmentation information and the part-of-speech tagging result, word groups are extracted from the first evaluation information to obtain the first word group set corresponding to the first evaluation information;

[0152] Obtain the number of times the first word group set appears in the historical evaluation information published by the target user, and calculate the sentiment value of the first word group corresponding to each first word group set;

[0153] The first evaluation sentiment value of the first evaluation information is obtained by calculating the sentiment values ​​of each of the first word groups.

[0154] In some feasible implementations, the historical evaluation information includes at least positive and negative evaluation information, and the vector determination module 203 is specifically used for:

[0155] Obtain the first occurrence count of the first phrase set in the positive review information;

[0156] Obtain the second occurrence count of the first phrase set in the negative review information;

[0157] The sentiment value of the first phrase corresponding to each set of the first phrases is obtained by calculating the first occurrence count and the second occurrence count using the following formula:

[0158]

[0159] Wherein, N1 is the first occurrence count, and N2 is the second occurrence count; if either N1 or N2 is 0, then the sentiment value of the corresponding first word set is not calculated.

[0160] In some feasible implementations, the vector determination module 203 is specifically used for:

[0161] The second evaluation information is segmented into words to obtain the second segmentation information;

[0162] Part-of-speech tagging is performed on the second word segmentation information to obtain the part-of-speech tagging result of the second word segmentation information;

[0163] Using the second word segmentation information and the part-of-speech tagging result, word groups are extracted from the second evaluation information to obtain the second word group set corresponding to the second evaluation information;

[0164] Obtain the number of times the second word set appears in the historical evaluation information published by the target user, and calculate the sentiment value of the second word set corresponding to each second word set;

[0165] The second evaluation sentiment value of the second evaluation information is obtained by calculating the sentiment values ​​of each of the second phrases.

[0166] In some feasible implementations, the historical evaluation information includes at least positive and negative evaluation information, and the vector determination module 203 is specifically used for:

[0167] Obtain the third occurrence count of the second phrase set in the positive review information;

[0168] Obtain the fourth occurrence count of the second phrase set in the negative review information;

[0169] The sentiment value of the second phrase corresponding to each set of the second phrase is obtained by calculating the third occurrence count and the fourth occurrence count using the following formula:

[0170]

[0171] Wherein, N1 is the third occurrence count, and N2 is the fourth occurrence count; if either N1 or N2 is 0, then the sentiment value of the corresponding second word set is not calculated.

[0172] In some feasible implementations, the vector determination module 203 is specifically used for:

[0173] The average of the first evaluation sentiment values ​​of each of the first evaluation information is taken as the sentiment value of the target benefit product, and the average of the second evaluation sentiment values ​​of each of the second evaluation information is taken as the sentiment value of the target product.

[0174] Obtain the equity product characteristics of the target equity product;

[0175] The first emotional feature vector corresponding to the target equity product is obtained by concatenating the equity product features, product category and emotional value of the target equity product.

[0176] The product category and the emotional value of the target product are concatenated to obtain the second emotional feature vector corresponding to the target product.

[0177] In some feasible implementations, the user rights parameters include at least the number of rights browsing sessions, transaction amount, and number of transactions. The product prediction module 204 is specifically used for:

[0178] The number of times the rights and interests are viewed, the transaction amount, the number of transactions, the first sentiment feature vector, and the second sentiment feature vector are output into a preset rights and interests prediction model to obtain the target rights and interests product.

[0179] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0180] In addition, embodiments of the present invention also provide an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the various processes of the above-described user preference prediction method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0181] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the above-described user preference prediction method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0182] Figure 3 A schematic diagram of the hardware structure of an electronic device for implementing various embodiments of the present invention.

[0183] The electronic device 300 includes, but is not limited to, components such as: a radio frequency unit 301, a network module 302, an audio output unit 303, an input unit 304, a sensor 305, a display unit 306, a user input unit 307, an interface unit 308, a memory 309, a processor 310, and a power supply 311. Those skilled in the art will understand that the electronic device structure involved in the embodiments of the present invention does not constitute a limitation on the electronic device. An electronic device may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. In the embodiments of the present invention, the electronic device includes, but is not limited to, mobile phones, tablet computers, laptop computers, PDAs, in-vehicle terminals, wearable devices, and pedometers.

[0184] It should be understood that, in this embodiment of the invention, the radio frequency unit 301 can be used for receiving and transmitting signals during information transmission or calls. Specifically, it receives downlink data from the base station and processes it with the processor 310; additionally, it transmits uplink data to the base station. Typically, the radio frequency unit 301 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low-noise amplifier, a duplexer, etc. Furthermore, the radio frequency unit 301 can also communicate with networks and other devices through a wireless communication system.

[0185] The electronic device provides users with wireless broadband internet access through the network module 302, such as helping users send and receive emails, browse web pages, and access streaming media.

[0186] The audio output unit 303 can convert audio data received by the radio frequency unit 301 or the network module 302 or stored in the memory 309 into audio signals and output them as sound. Furthermore, the audio output unit 303 can also provide audio output related to specific functions performed by the electronic device 300 (e.g., call signal reception sound, message reception sound, etc.). The audio output unit 303 includes a speaker, a buzzer, and a receiver, etc.

[0187] Input unit 304 is used to receive audio or video signals. Input unit 304 may include a graphics processing unit (GPU) 3041 and a microphone 3042. The GPU 3041 processes image data of still images or videos acquired by an image capture device (such as a camera) in video capture mode or image capture mode. The processed image frames can be displayed on display unit 306. The image frames processed by GPU 3041 can be stored in memory 309 (or other storage media) or transmitted via radio frequency unit 301 or network module 302. Microphone 3042 can receive sound and process such sound into audio data. The processed audio data can be converted into a format that can be transmitted to a mobile communication base station via radio frequency unit 301 in telephone call mode.

[0188] The electronic device 300 also includes at least one sensor 305, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor includes an ambient light sensor and a proximity sensor. The ambient light sensor can adjust the brightness of the display panel 3061 according to the ambient light level, and the proximity sensor can turn off the display panel 3061 and / or backlight when the electronic device 300 is moved to the ear. As a type of motion sensor, an accelerometer sensor can detect the magnitude of acceleration in various directions (generally three axes). When stationary, it can detect the magnitude and direction of gravity and can be used to identify the posture of the electronic device (such as landscape / portrait switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc. The sensor 305 may also include a fingerprint sensor, pressure sensor, iris sensor, molecular sensor, gyroscope, barometer, hygrometer, thermometer, infrared sensor, etc., which will not be described in detail here.

[0189] The display unit 306 is used to display information input by the user or information provided to the user. The display unit 306 may include a display panel 3061, which may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0190] User input unit 307 can be used to receive input numerical or character information, and to generate key signal inputs related to user settings and function control of electronic devices. Specifically, user input unit 307 includes a touch panel 3071 and other input devices 3072. Touch panel 3071, also known as a touch screen, can collect touch operations performed by the user on or near it (such as operations performed by the user using a finger, stylus, or any suitable object or accessory on or near touch panel 3071). Touch panel 3071 may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch position and the signal generated by the touch operation, and transmits the signal to the touch controller; the touch controller receives touch information from the touch detection device, converts it into touch point coordinates, and sends it to processor 310, which receives and executes commands from processor 310. In addition, touch panel 3071 can be implemented using various types such as resistive, capacitive, infrared, and surface acoustic wave. In addition to touch panel 3071, user input unit 307 may also include other input devices 3072. Specifically, other input devices 3072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, joysticks, etc., which will not be described in detail here.

[0191] Furthermore, the touch panel 3071 can cover the display panel 3061. When the touch panel 3071 detects a touch operation on or near it, it transmits the information to the processor 310 to determine the type of touch event. Subsequently, the processor 310 provides corresponding visual output on the display panel 3061 according to the type of touch event. It is understood that in one embodiment, the touch panel 3071 and the display panel 3061 are implemented as two independent components to realize the input and output functions of the electronic device. However, in some embodiments, the touch panel 3071 and the display panel 3061 can be integrated to realize the input and output functions of the electronic device. The specific implementation is not limited here.

[0192] Interface unit 308 serves as an interface for connecting external devices to electronic device 300. For example, external devices may include a wired or wireless headphone port, an external power supply (or battery charger) port, a wired or wireless data port, a memory card port, a port for connecting a device with an identification module, an audio input / output (I / O) port, a video I / O port, a headphone port, and so on. Interface unit 308 can be used to receive input from external devices (e.g., data, power, etc.) and transmit the received input to one or more components within electronic device 300, or it can be used to transmit data between electronic device 300 and external devices.

[0193] The memory 309 can be used to store software programs and various data. The memory 309 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function (such as sound playback, image playback, etc.), etc.; the data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). Furthermore, the memory 309 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0194] The processor 310 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 309, and by calling data stored in the memory 309, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. The processor 310 may include one or more processing units; preferably, the processor 310 may integrate an application processor and a modem processor. The application processor mainly handles the operating system, user interface, and applications, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 310.

[0195] The electronic device 300 may also include a power supply 311 (such as a battery) for supplying power to various components. Preferably, the power supply 311 can be logically connected to the processor 310 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system.

[0196] In addition, the electronic device 300 includes some functional modules not shown, which will not be described in detail here.

[0197] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0198] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0199] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

[0200] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0201] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0202] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0203] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0204] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0205] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0206] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for predicting user preferences, characterized in that, include: Obtain the target users' initial evaluation information on the target benefit products they have purchased in the past; Identify target products that belong to the same category as the target benefit product, and obtain the second evaluation information of the target user on the target product; Calculate the first evaluation sentiment value corresponding to the first evaluation information and the second evaluation sentiment value corresponding to the second evaluation information, and integrate the first evaluation sentiment value and the second evaluation sentiment value to obtain a first sentiment feature vector corresponding to the target benefit product and a second sentiment feature vector corresponding to the target product; Obtain the user rights parameters of the target user, and perform product prediction based on the first sentiment feature vector, the second sentiment feature vector, and the user rights parameters to obtain the target rights product; The step of integrating the first and second evaluation sentiment values ​​to obtain a first sentiment feature vector corresponding to the target benefit product and a second sentiment feature vector corresponding to the target product includes: The average of the first evaluation sentiment values ​​of each of the first evaluation information is taken as the sentiment value of the target benefit product, and the average of the second evaluation sentiment values ​​of each of the second evaluation information is taken as the sentiment value of the target product. Obtain the equity product characteristics of the target equity product; The first emotional feature vector corresponding to the target equity product is obtained by concatenating the equity product features, product category and emotional value of the target equity product. The product category and the emotional value of the target product are concatenated to obtain the second emotional feature vector corresponding to the target product.

2. The method according to claim 1, characterized in that, The calculation of the first evaluation sentiment value corresponding to the first evaluation information includes: The first evaluation information is segmented into words to obtain the first segmentation information; Part-of-speech tagging is performed on the first segmented word information to obtain the part-of-speech tagging result of the first segmented word information; Using the first word segmentation information and the part-of-speech tagging result, word groups are extracted from the first evaluation information to obtain the first word group set corresponding to the first evaluation information; Obtain the number of times the first word group set appears in the historical evaluation information published by the target user, and calculate the sentiment value of the first word group corresponding to each first word group set; The first evaluation sentiment value of the first evaluation information is obtained by calculating the sentiment values ​​of each of the first word groups.

3. The method according to claim 2, characterized in that, The historical evaluation information includes at least positive and negative evaluation information. The step of obtaining the frequency of occurrence of the first word set in the historical evaluation information posted by the target user, and calculating the sentiment value of the first word set corresponding to each first word set, includes: Obtain the first occurrence count of the first phrase set in the positive review information; Obtain the second occurrence count of the first phrase set in the negative review information; The sentiment value of the first phrase corresponding to each first phrase set is obtained by calculating the first occurrence count and the second occurrence count using the following formula: Wherein, N1 is the first occurrence count, and N2 is the second occurrence count; if either N1 or N2 is 0, then the sentiment value of the corresponding first word set is not calculated.

4. The method according to claim 1, characterized in that, The calculation of the second evaluation sentiment value corresponding to the second evaluation information includes: The second evaluation information is segmented into words to obtain the second segmentation information; Part-of-speech tagging is performed on the second word segmentation information to obtain the part-of-speech tagging result of the second word segmentation information; Using the second word segmentation information and the part-of-speech tagging result, word groups are extracted from the second evaluation information to obtain the second word group set corresponding to the second evaluation information; Obtain the number of times the second word set appears in the historical evaluation information published by the target user, and calculate the sentiment value of the second word set corresponding to each second word set; The second evaluation sentiment value of the second evaluation information is obtained by calculating the sentiment values ​​of each of the second phrases.

5. The method according to claim 4, characterized in that, The historical evaluation information includes at least positive and negative evaluation information. The step of obtaining the frequency of occurrence of the second word set in the historical evaluation information posted by the target user, and calculating the sentiment value of the second word set corresponding to each second word set, includes: Obtain the third occurrence count of the second phrase set in the positive review information; Obtain the fourth occurrence count of the second phrase set in the negative review information; The sentiment value of the second phrase corresponding to each set of the second phrase is obtained by calculating the third occurrence count and the fourth occurrence count using the following formula: Wherein, N1 is the third occurrence count, and N2 is the fourth occurrence count; if either N1 or N2 is 0, then the sentiment value of the corresponding second word set is not calculated.

6. The method according to claim 1, characterized in that, The user benefit parameters include at least the number of benefit views, transaction amount, and number of transactions. The step of predicting the target benefit product based on the first sentiment feature vector, the second sentiment feature vector, and the user benefit parameters includes: The number of times the rights and interests are viewed, the transaction amount, the number of transactions, the first sentiment feature vector, and the second sentiment feature vector are output into a preset rights and interests prediction model to obtain the target rights and interests product.

7. A user preference prediction device, characterized in that, include: The product acquisition module is used to obtain the target users' first evaluation information on the target benefit products they have purchased in the past. The evaluation information acquisition module is used to determine the target product that is of the same category as the target benefit product, and to acquire the second evaluation information of the target user on the target product; The vector determination module is used to calculate the first evaluation sentiment value corresponding to the first evaluation information and the second evaluation sentiment value corresponding to the second evaluation information, and to integrate the first evaluation sentiment value and the second evaluation sentiment value to obtain a first sentiment feature vector corresponding to the target benefit product and a second sentiment feature vector corresponding to the target product. The product prediction module is used to obtain the user rights parameters of the target user, and to perform product prediction based on the first sentiment feature vector, the second sentiment feature vector and the user rights parameters to obtain the target rights product. Specifically, the vector determination module is used for: The average of the first evaluation sentiment values ​​of each of the first evaluation information is taken as the sentiment value of the target benefit product, and the average of the second evaluation sentiment values ​​of each of the second evaluation information is taken as the sentiment value of the target product. Obtain the equity product characteristics of the target equity product; The first emotional feature vector corresponding to the target equity product is obtained by concatenating the equity product features, product category and emotional value of the target equity product. The product category and the emotional value of the target product are concatenated to obtain the second emotional feature vector corresponding to the target product.

8. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes a program stored in the memory, it implements the method as described in any one of claims 1-6.

9. A computer-readable storage medium having instructions stored thereon that, when executed by one or more processors, cause the processors to perform the method as described in any one of claims 1-6.

Images