Commodity display method and device, computer readable storage medium and terminal

Abstract

The invention discloses a commodity display method and device, a computer readable storage medium and a terminal. The method comprises the steps of obtaining search information; performing word segmentation on the search information to determine a search word set; determining the matching degree of each existing commodity by adopting the search word set; determining operation information of each existing commodity, wherein the operation information is at least used for representing whether the existing commodity has a preset operation attribute; weighting operation is carried out at least based on the matching degree and the operation information, display values of the existing commodities are obtained one by one, a preset first weight value is adopted as the weight of the matching degree of each existing commodity, and an adjustable second weight value is adopted as the weight of the operation information of each existing commodity; and sorting and displaying at least one part of the existing commodities based on the display values. According to the method, information of more dimensions can be combined, and the comprehensiveness and systematicness of commodity display are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a commodity display method and device, a computer-readable storage medium, and a terminal. Background Art

[0002] With the rapid development of e-commerce, online product selection and purchasing models are gaining more and more attention, and mobile online shopping mall platforms have begun to be established to attract more online users.

[0003] However, the accuracy of existing shopping mall product recommendations is insufficient, which seriously affects users' search experience and thus sales.

[0004] In an existing product display method, a full-text search is performed in the database for the search terms entered by the user, and the corresponding products are matched. Finally, the search results are aggregated and displayed to the user. However, this display method analyzes and considers few dimensions, and only mechanically makes recommendations based on the user's input. The actual recommendations lack comprehensiveness and systematicness.

[0005] There is an urgent need for a product display method that can combine information from more dimensions to improve the comprehensiveness and systematicness of product display. Summary of the invention

[0006] The technical problem solved by the present invention is to provide a commodity display method and device, a computer-readable storage medium, and a terminal, which can combine information of more dimensions to improve the comprehensiveness and systematicness of commodity display.

[0007] To solve the above technical problems, an embodiment of the present invention provides a commodity display method, comprising: obtaining search information; segmenting the search information to determine a search word set; using the search word set to determine the matching degree of each existing commodity; determining the operation information of each existing commodity, wherein the operation information is at least used to indicate whether the existing commodity has a preset operation attribute; performing a weighted operation based at least on the matching degree and the operation information to obtain the display value of the existing commodities one by one, wherein a preset first weight value is used as the weight of the matching degree of each existing commodity, and an adjustable second weight value is used as the weight of the operation information of each existing commodity; based on the display value, at least a portion of the existing commodities are sorted and displayed.

[0008] Optionally, the operation information is selected from one or more of the following: whether each existing product is a self-operated product; whether each existing product belongs to a preset supplier.

[0009] Optionally, a lower limit of an adjustable range of the second weight value is greater than or equal to the first weight value.

[0010] Optionally, the search information is segmented to determine a search word set, including: preprocessing the search information to obtain a preprocessing result; using an inverse maximum matching algorithm to match each word in the preprocessing result in a word database until a matching word is found for each word or only a single word remains; and constructing the search word set, which includes each matching word and its replacement word.

[0011] Optionally, the search information is preprocessed to obtain a preprocessing result, including: filtering the search information by preset words, wherein the preset words include one or more of the following: modal particles, auxiliary words, garbled characters, sensitive words and stop words; and segmenting the filtered search information to obtain the preprocessing result.

[0012] Optionally, the replacement word for each matching word is selected from: a synonym of the matching word and / or a near synonym of the matching word.

[0013] Optionally, the search word set is used to determine the matching degree of each existing product, including: constructing a data vector to be searched based on a data vector of a preset dimension using the search word set, wherein data at at least a portion of the positions in the data vector of the preset dimension are used to indicate at least a portion of the words in the word database in a one-to-one correspondence; and using a cosine similarity algorithm to compare the data vector to be searched with the data vector of each existing product one by one to determine the matching degree of each existing product.

[0014] Optionally, based on the data vector of preset dimension, the search word set is used to construct the data vector to be searched, including: traversing the data vector of preset dimension, assigning the position of the word in the search word set to the first preset value, and assigning the remaining positions to the second preset value, so as to obtain the data vector to be searched.

[0015] Optionally, before comparing the data vector to be searched with the data vector of each existing product one by one, the method also includes: performing word segmentation on the product information of each existing product to obtain a product word set and the word database for each existing product; for each existing product, traversing the data vectors of the preset dimension, assigning the position of the word in the product word set to a first preset value, and assigning the remaining positions to a second preset value, so as to obtain the data vector of each existing product.

[0016] Optionally, the data vector of the preset dimension is a single-row and N-column vector or an N-row and single-column vector; wherein N is used to represent the preset dimension, and N is greater than or equal to the number of words in the word database.

[0017] Optionally, the performing of weighted operations at least based on the matching degree and the operation information to obtain the display values ​​of the existing commodities one by one includes: performing weighted operations based on the matching degree, the operation information and the first parameter group to obtain the display values ​​of the existing commodities one by one; wherein the first parameter group has a preset third weight value, and the third weight value is smaller than the first weight value, and smaller than the second weight value.

[0018] Optionally, the first parameter group includes one or more of the following: a level value belonging to the number of historical searches for each existing product; a level value belonging to the shelf time of each existing product; wherein, the more times the existing product is searched, the greater the level value belonging to the number of historical searches; and the shorter the shelf time of the existing product, the greater the level value belonging to the shelf time.

[0019] Optionally, the weighted operation is performed at least based on the matching degree and the operation information to obtain the display value of the existing goods one by one, including: performing a weighted operation based on the matching degree, the operation information, the first parameter group and the second parameter group to obtain the display value of the existing goods one by one; wherein the second parameter group has a preset fourth weight value, and the fourth weight value is less than the third weight value.

[0020] Optionally, the second parameter group includes one or more of the following: the level value to which the historical click-through rate of each existing product belongs; the level value to which the historical sales volume of each existing product belongs; the promotion value of each existing product; wherein, the more times the existing product is clicked, the greater the level value to which the historical click-through rate belongs; the more historical sales volume of the existing product, the greater the level value to which the historical sales volume belongs; if the existing product has promotion information, the promotion value is a first preset value; if the existing product does not have promotion information, the promotion value is a second preset value, and the first preset value is greater than the second preset value.

[0021] To solve the above technical problems, an embodiment of the present invention provides a commodity display device, comprising: an acquisition module, used to acquire search information; a word segmentation module, used to segment the search information to determine a search word set; a matching degree determination module, used to use the search word set to determine the matching degree of each existing commodity; an operation determination module, used to determine the operation information of each existing commodity, the operation information at least used to indicate whether the existing commodity participates in an operation activity; a weighting module, used to perform weighted operations based on at least the matching degree and the operation information to obtain display values ​​of the existing commodities one by one, wherein a preset first weight value is used as the weight of the matching degree of each existing commodity, and an adjustable second weight value is used as the weight of the operation information of each existing commodity; a display module, used to sort and display at least a part of the existing commodities based on the display value.

[0022] In order to solve the above technical problem, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above commodity display method are executed.

[0023] To solve the above technical problems, an embodiment of the present invention provides a terminal, including a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor executes the steps of the above commodity display method when running the computer program.

[0024] Compared with the prior art, the technical solution of the embodiment of the present invention has the following beneficial effects:

[0025] In an embodiment of the present invention, by segmenting the search information and using the search word set to determine the matching degree of each existing product, the products that match the search information input by the user can be displayed first. On this basis, by determining the operating information of each existing product, a weighted operation is performed at least based on the matching degree and the operating information to obtain the display value of the existing products one by one, and at least a part of the existing products are sorted and displayed. This can increase the dimension of analysis, consider more dimensions on the basis of the search information input by the user and the operating information of the product, and effectively improve the comprehensiveness and systematicness of the display. Furthermore, by using an adjustable second weight value as the weight of the operating information of each existing product, the corresponding weight can be flexibly configured according to the diversity of operating activities, such as priority display or delayed display, thereby further improving the effectiveness and accuracy of the display.

[0026] Furthermore, the operation information is selected from one or more of the following: whether each existing product is a self-operated product; whether each existing product belongs to a preset supplier, so that for the situation where the product is self-operated by the mall, the product belongs to a preset supplier, and the product participates in a preset marketing activity, adjustable weights can be used to adjust the display situation, such as priority display or delayed display, so as to further improve the effectiveness and accuracy of the display.

[0027] Furthermore, the lower limit of the adjustable range of the second weight value is greater than or equal to the first weight value, thereby further improving the influence of operational information in the process of calculating the display value of existing products, making more obvious adjustments to the display position (or display order) of products with preset operational attributes, and further improving the effectiveness of the display.

[0028] Furthermore, the step of segmenting the search information to determine the search word set includes: preprocessing the search information to obtain a preprocessing result; using an inverse maximum matching algorithm to match each word in the preprocessing result in a word database until a matching word is found for each word or only a single word remains; constructing the search word set, which includes each matching word and its replacement word, so that based on the search information input by the user, the product can be associated by selecting the replacement word, thereby effectively increasing the search scope and displaying the products that the user may be interested in to a greater extent.

[0029] Furthermore, the matching degree of each existing product is determined by using the search word set, including: constructing a data vector to be searched using the search word set according to a data vector of a preset dimension, wherein the data of at least a portion of the positions in the data vector of the preset dimension are used to indicate at least a portion of the words in the word database in a one-to-one correspondence; using a cosine similarity algorithm, comparing the data vector to be searched with the data vector of each existing product one by one to determine the matching degree of each existing product. By using the above scheme, a data vector of a preset dimension can be constructed, and the data at each position indicates the words in the word database in a one-to-one correspondence, which can be used as a comparison basis for the cosine similarity algorithm, thereby further improving the accuracy of determining the matching degree in the process of determining the matching degree of each existing product.

[0030] Further, the data vectors of the preset dimension are traversed, and the positions of the words in the search word set are assigned the first preset value, and the remaining positions are assigned the second preset value, so as to obtain the data vector to be searched. By adopting the above scheme, by assigning the positions of the words in the search word set the first preset value, the data vector to be searched is also a data vector of the same dimension, thereby further reducing the computational complexity and improving the accuracy of the comparison result in the process of adopting the cosine similarity algorithm.

[0031] Furthermore, the product information of each existing product is segmented to obtain a product word set of each existing product and the word database. For each existing product, the data vector of the preset dimension is traversed, and the position of the word in the product word set is assigned a first preset value, and the remaining positions are assigned a second preset value to obtain the data vector of each existing product. By adopting the above scheme, by assigning the position of the word in the product word set to the first preset value, the data vector of each existing product and the data vector to be searched are both data vectors of the same dimension, thereby further reducing the computational complexity and improving the accuracy of the comparison result in the process of adopting the cosine similarity algorithm.

[0032] Furthermore, the data vector of the preset dimension is a single-row and N-column vector or an N-row and single-column vector; wherein N is used to represent the preset dimension, and N is greater than or equal to the number of words in the word database. In one case, the data vector can be set to correspond one-to-one with the words in the word database, and in another case, some vacant positions can be reserved in the data vector, so that when new words are added to the word database, the data vectors of existing products that have been configured will not be affected, thereby effectively reducing the frequency of data processing and reducing maintenance costs.

[0033] Furthermore, the weighted operation is performed at least based on the matching degree and the operation information to obtain the display value of the existing products one by one, including: weighted operation is performed based on the matching degree, the operation information and the first parameter group to obtain the display value of the existing products one by one; wherein the first parameter group has a preset third weight value, and the third weight value is less than the first weight value and less than the second weight value. The above scheme can be combined with the first parameter group to further increase the dimension of analysis, and while ensuring the importance of the display of matching degree and operation information, more dimensions can be considered to effectively improve the comprehensiveness and systematicness of the display.

[0034] Furthermore, the weighted operation is performed at least based on the matching degree and the operation information to obtain the display value of the existing goods one by one, including: weighted operation is performed based on the matching degree, the operation information, the first parameter group and the second parameter group to obtain the display value of the existing goods one by one; wherein the second parameter group has a preset fourth weight value, and the fourth weight value is less than the third weight value. The above scheme can be combined with the second parameter group to further increase the dimension of analysis, and while ensuring the matching degree, the operation information, and the display importance of the first parameter group, more dimensions can be considered to effectively improve the comprehensiveness and systematicness of the display. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a flow chart of a commodity display method in an embodiment of the present invention;

[0036] Figure 2 yes Figure 1 A flowchart of a specific implementation of step S12;

[0037] Figure 3 It is a structural schematic diagram of a commodity display device in an embodiment of the present invention. DETAILED DESCRIPTION

[0038] As mentioned above, the accuracy of existing mall product recommendations is insufficient, which seriously affects users' search experience and thus sales.

[0039] In an existing product display method, a full-text search is performed in the database for the search terms entered by the user, the corresponding products are matched, and finally the search results are aggregated and displayed to the user.

[0040] After research, it was found that in the prior art, the search is only based on the information input by the user. For example, if the user inputs a certain type of product of a certain brand, the brand name and product name will be searched, and the matching products will be displayed to the user, such as products with various quantity units and from various suppliers. However, this display method has few dimensions for analysis and consideration, and only mechanically recommends according to the user's input, and the actual recommendation is not comprehensive and systematic enough.

[0041] In an embodiment of the present invention, by segmenting the search information and using the search word set to determine the matching degree of each existing product, the products that match the search information input by the user can be displayed first. On this basis, by determining the operating information of each existing product, a weighted operation is performed at least based on the matching degree and the operating information to obtain the display value of the existing products one by one, and at least a part of the existing products are sorted and displayed. This can increase the dimension of analysis, consider more dimensions on the basis of the search information input by the user and the operating information of the product, and effectively improve the comprehensiveness and systematicness of the display. Furthermore, by using an adjustable second weight value as the weight of the operating information of each existing product, the corresponding weight can be flexibly configured according to the diversity of operating activities, such as priority display or delayed display, thereby further improving the effectiveness and accuracy of the display.

[0042] In order to make the above-mentioned objects, features and beneficial effects of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0043] Reference Figure 1 , Figure 1 1 is a flow chart of a commodity display method in an embodiment of the present invention. The commodity display method may include steps S11 to S16:

[0044] Step S11: obtaining search information;

[0045] Step S12: segmenting the search information to determine a search word set;

[0046] Step S13: using the search word set to determine the matching degree of each existing product;

[0047] Step S14: determining the operation information of each existing commodity, where the operation information is at least used to indicate whether the existing commodity has a preset operation attribute;

[0048] Step S15: performing a weighted calculation based at least on the matching degree and the operation information to obtain display values ​​of the existing products one by one, wherein a preset first weight value is used as the weight of the matching degree of each existing product, and an adjustable second weight value is used as the weight of the operation information of each existing product;

[0049] Step S16: Based on the display value, at least a portion of the existing products are sorted and displayed.

[0050] In the specific implementation of step S11, the search information may be externally input information, such as information input by a user, and may also include the name, brand, attribute, quantity, etc. of the product.

[0051] The name of the product is used to distinguish the products, for example, it may include "milk", "coffee", etc.

[0052] The brand of a product is used to distinguish the source of the product. For example, it may include the supplier of the product, the origin of the product, etc.

[0053] The attributes of a product are used to describe the characteristics of the product, and may include, for example, "peach flavor", "non-GMO", "constant temperature and warmth", etc.

[0054] The quantity of a product is used to indicate the amount contained in a single product, for example, it may include "200ml", "280g", etc.

[0055] In the specific implementation of step S12, a conventional word segmentation method may be used, such as searching and retaining each noun in the search information as a search word set; an improved word segmentation method in an embodiment of the present invention may also be used to improve the word segmentation effect.

[0056] Reference Figure 2 , Figure 2 yes Figure 1 The step of segmenting the search information to determine the search word set may include steps S21 to S23, each of which is described below.

[0057] In step S21, the search information is preprocessed to obtain a preprocessing result.

[0058] Among them, preprocessing can be performed based on one or more of the following algorithms: dictionary bidirectional maximum matching algorithm, soft decision (VITERBI) algorithm, HMM algorithm and conditional random field algorithm (Conditional random field algorithm, CRF algorithm) algorithm

[0059] Furthermore, the step of preprocessing the search information to obtain the preprocessing result may include: filtering the search information by preset words, wherein the preset words include one or more of the following: modal particles, auxiliary words, garbled characters, sensitive words and stop words; and segmenting the filtered search information to obtain the preprocessing result.

[0060] In the embodiment of the present invention, modal particles, auxiliary words, garbled characters, sensitive words and stop words are of little use to the search and will affect the search results. By removing these words and performing word segmentation on the remaining content, the search effectiveness and accuracy can be improved.

[0061] It should be noted that in another specific implementation of the embodiment of the present invention, the corpus may also be processed by word retention, for example, retaining one or more of the following: nouns, adjectives, and quantifiers, to obtain a preprocessing result.

[0062] In the embodiment of the present invention, nouns, adjectives and quantifiers are more useful for searching and will affect the search results. By retaining the words and performing word segmentation on the remaining content, the search effectiveness and accuracy can be improved.

[0063] In step S22, a reverse maximum matching algorithm is used to match each word in the preprocessing result in the word database until a matching word is found for each word or only a single word remains.

[0064] Specifically, you can first determine whether the word is in the word database. If so, treat it as a single matching word. If not, reduce a single character from the right, and then determine whether the reduced word is in the word database. If so, treat it as a single matching word. If not, continue to reduce a single character from the right, and then determine whether the reduced word is in the word database. This cycle is repeated until only a single word is left.

[0065] It should be noted that "a single word is left" may include the case of "a single character is left".

[0066] In the embodiment of the present invention, by adopting the reverse maximum matching algorithm, the accuracy and comprehensiveness of searching for matching words can be improved.

[0067] In step S23, the search word set is constructed, and the search word set includes each matching word and its replacement word.

[0068] The replacement words may be generated based on one or more of the following methods: based on an ibmmodel translation alignment model, based on a seq2seq neural network translation model, based on a word embedding vector similarity calculation model, or based on a rule-based statistical model.

[0069] Furthermore, the replacement word for each matching word may be selected from: a synonym of the matching word and / or a near synonym of the matching word.

[0070] Among them, synonyms can be used to represent words with the same lexical meaning, also called synonyms, such as "mercury" and "mercury", "kiwi fruit" and "kiwi fruit", etc.

[0071] Synonyms can be used to represent words with similar lexical meanings, such as "edible oil" and "peanut oil", "rapeseed oil", "soybean oil", etc.

[0072] In the embodiment of the present invention, by using the synonyms of the matching word and / or the near synonyms of the matching word as replacement words, the search scope can be increased and the effectiveness of searching based on the user's search information can be improved.

[0073] In an embodiment of the present invention, the step of segmenting the search information to determine the search word set includes: preprocessing the search information to obtain a preprocessing result; using an inverse maximum matching algorithm to match each word in the preprocessing result in a word database until a matching word is found for each word or only a single word remains; constructing the search word set, which includes each matching word and its replacement word, so that based on the search information input by the user, the product can be associated by selecting the replacement word, thereby effectively increasing the search scope and displaying the products that the user may be interested in to a greater extent.

[0074] In the specific implementation of step S13, the matching degree can be determined in a conventional and appropriate manner, for example, the number of words in the search word set that match the word database can be counted, and the more words that match, the higher the matching degree is. The improved method for determining the matching degree in the embodiment of the present invention can also be used to improve the accuracy and effect of the determination.

[0075] Furthermore, the step of using the search word set to determine the matching degree of each existing product may include: constructing a data vector to be searched using the search word set based on a data vector of a preset dimension, wherein data at at least a portion of the positions in the data vector of the preset dimension are used to indicate at least a portion of the words in the word database in a one-to-one correspondence; using a cosine similarity algorithm, comparing the data vector to be searched with the data vector of each existing product one by one to determine the matching degree of each existing product.

[0076] Furthermore, before comparing the data vector to be searched with the data vector of each existing product one by one, the method may also include: performing word segmentation on the product information of each existing product to obtain a product word set and the word database for each existing product; for each existing product, traversing the data vector of the preset dimension, assigning a first preset value to the position of the word in the product word set, and assigning a second preset value to the remaining positions, so as to obtain the data vector of each existing product.

[0077] Among them, the word database can be pre-built and continuously maintained, and the word database can contain the word segmentation information of each existing product. The word segmentation information of each existing product includes the smallest word, the phrase obtained by splicing multiple words, and can also include the attribute information of each word and at least part of the phrase, such as the name, brand, attribute, quantity, etc. of the product to which the word belongs.

[0078] In a non-limiting specific embodiment, taking a product library with 10,000 existing products as an example, after all the existing products are segmented, a total of 1,000 segmentation phrases are obtained as the key of the vector, that is, the preset dimension of the data vector is 1,000.

[0079] Specifically, the following 1000 word segmentation vector key sets can be formed: [milk, coffee, apple, original flavor, 200ml, 5L, ...].

[0080] Then, each existing product can be assigned a value to obtain a data vector for each existing product.

[0081] For example, after assigning values ​​to “Original milk 200ml”, the data vector obtained is [1, 0, 0, 1, 1, 0, …].

[0082] In an embodiment of the present invention, the product information of each existing product is segmented to obtain a product word set and the word database of each existing product. For each existing product, the data vector of the preset dimension is traversed, and the position of the word in the product word set is assigned a first preset value, and the remaining positions are assigned a second preset value to obtain the data vector of each existing product. By adopting the above scheme, by assigning the position of the word in the product word set to the first preset value, the data vector of each existing product and the data vector to be searched are both data vectors of the same dimension, thereby further reducing the computational complexity and improving the accuracy of the comparison result in the process of adopting the cosine similarity algorithm.

[0083] Furthermore, based on the data vector of preset dimension, the step of constructing the data vector to be searched using the search word set may include: traversing the data vector of preset dimension, assigning the position of the word in the search word set to a first preset value, and assigning the remaining positions to a second preset value, so as to obtain the data vector to be searched.

[0084] Specifically, the search word set is also assigned values, and the data vector to be searched can be obtained.

[0085] For example, after assigning values ​​to “coffee milk”, the data vector obtained is [1, 1, 0, 0, 0, 0, …].

[0086] In an embodiment of the present invention, by traversing the data vector of the preset dimension, the position of the word in the search word set is assigned a first preset value, and the remaining positions are assigned a second preset value, so as to obtain the data vector to be searched. By adopting the above scheme, by assigning the position of the word in the search word set to the first preset value, the data vector to be searched is also a data vector of the same dimension, thereby further reducing the computational complexity and improving the accuracy of the comparison result in the process of adopting the cosine similarity algorithm.

[0087] Then, the cosine similarity algorithm may be used to compare the data vector to be searched with the data vector of each existing product one by one to determine the matching degree of each existing product.

[0088] Specifically, take the case where there are 10,000 existing products in the product library, each product has a data vector of 1000 dimensions, and the data vectors of 10,000 existing products are obtained as an example. The data vectors of 10,000 existing products can be regarded as vector A, which is similar to:

[0089] [1,1,0,0,1,......];

[0090] [1,0,0,0,1,......];

[0091] …

[0092] The data vector to be searched may also be recorded as vector B. Since the dimension of the data vector to be searched is also a preset dimension, which is consistent with the dimension of the data vector of the existing products, the cosine similarity between the data vector to be searched and the data vector of each existing product may be calculated respectively.

[0093] Furthermore, the following formula can be used to determine the matching degree between the data vector to be searched and each existing product:

[0094]

[0095] Among them, A is used to represent vector A, A iis used to represent the i-th data in vector A, B is used to represent vector B, B i It is used to represent the i-th data in vector B, and n is used to represent the number of data in vector A and vector B.

[0096] Take n=4, A vector is [1,1,0,1], B vector is [1,0,0,1] as an example, substitute into the above formula:

[0097]

[0098] It can be obtained that the matching degree between the data vector to be searched and the vector A is 0.63. It can be understood that the above matching degree can be obtained for each data vector of an existing product.

[0099] In an embodiment of the present invention, the search word set is used to determine the matching degree of each existing product, including: constructing a data vector to be searched using the search word set according to a data vector of a preset dimension, wherein the data of at least a portion of the positions in the data vector of the preset dimension are used to indicate at least a portion of the words in the word database in a one-to-one correspondence; using a cosine similarity algorithm, comparing the data vector to be searched with the data vector of each existing product one by one to determine the matching degree of each existing product. By adopting the above scheme, a data vector of a preset dimension can be constructed, and the data at each position indicates the words in the word database in a one-to-one correspondence, which can be used as a comparison basis for the cosine similarity algorithm, thereby further improving the accuracy of determining the matching degree in the process of determining the matching degree of each existing product.

[0100] Furthermore, the data vector of the preset dimension is a single-row and N-column vector or an N-row and single-column vector; wherein N is used to represent the preset dimension, and N is greater than or equal to the number of words in the word database.

[0101] Specifically, for the case of a single-row N-column vector, each word in the word database can have a corresponding column vector in the N column vectors based on a preset order. For the case of a single-column vector with N rows, each word in the word database can have a corresponding row vector in the N column vectors based on a preset order.

[0102] In an embodiment of the present invention, the data vector of the preset dimension is a single-row and N-column vector or an N-row and single-column vector; wherein N is used to represent the preset dimension, and N is greater than or equal to the number of words in the word database. In one case, the data vector can be set to correspond one-to-one with the words in the word database, and in another case, some vacant positions can be reserved in the data vector, so that when a new word is added to the word database, the data vectors of the existing products that have been configured will not be affected, thereby effectively reducing the frequency of data processing required and reducing maintenance costs.

[0103] In the specific implementation of step S14, the operation information of each existing commodity is determined, and the operation information is at least used to indicate whether the existing commodity has a preset operation attribute.

[0104] Specifically, a product operation information database can be built for existing products to store data on product operations, such as whether the product has participated in marketing activities or other promotional activities, the number of consumers covered, and other data.

[0105] It should be noted that, in the embodiment of the present invention, the operation information is at least used to indicate whether the existing commodity has a preset operation attribute.

[0106] The operation attribute may be used to indicate the supplier to which the existing goods belong.

[0107] Furthermore, the operation information is selected from one or more of the following: whether each existing product is a self-operated product; whether each existing product belongs to a preset supplier.

[0108] Among them, self-operated products can be used to indicate products operated by the mall itself, and preset suppliers can be used to indicate the source of products. In these two cases, the products can be displayed first to increase the probability of being noticed by users.

[0109] In an embodiment of the present invention, the operation information is selected from one or more of the following: whether each existing product is a self-operated product; whether each existing product belongs to a preset supplier, so that for the situation where the product is self-operated by the mall, the situation where the product belongs to a preset supplier, and the situation where the product participates in a preset marketing activity, an adjustable weight can be used to adjust the display situation, such as priority display or delayed display, so as to further improve the effectiveness and accuracy of the display.

[0110] In the specific implementation of step S15, a weighted operation is performed at least based on the matching degree and the operation information to obtain the display value of the existing products one by one, wherein a preset first weight value is used as the weight of the matching degree of each existing product, and an adjustable second weight value is used as the weight of the operation information of each existing product.

[0111] Specifically, the weighted calculation method in the embodiment of the present invention can be adopted, such as weighted average calculation to obtain the display value of existing goods to improve the effectiveness and accuracy of the display; other weighted calculation methods can also be adopted, such as weighted summation calculation or weighted median calculation method to obtain the display value of existing goods.

[0112] In an embodiment of the present invention, by adopting a preset first weight value as the weight of the matching degree of each existing commodity, a consistent weight can be configured for the existing commodities, thereby improving the fairness of display among the commodities, and by adopting an adjustable second weight value as the weight of the operation information of each existing commodity, the corresponding weight can be flexibly configured according to the diversity of operation activities, such as priority display or delayed display, thereby further improving the effectiveness and accuracy of the display.

[0113] Furthermore, the lower limit of the adjustable range of the second weight value may be greater than or equal to the first weight value.

[0114] Specifically, for example, for commodities with preset operational attributes, the operational information is set to a positive preset value, and for commodities with preset operational attributes, the operational information is set to a negative preset value. At this time, a larger second weight value is used, so that commodities with preset operational attributes can be more obviously given priority for display, and commodities without preset operational attributes can be more obviously delayed for display, or even not displayed.

[0115] As a non-limiting example, the weight value can be set to 1 for self-operated products and -1 for non-self-operated products, and a larger second weight value, such as 1000, can be set, so that the display position (or display order) of products with preset operational attributes can be more significantly adjusted.

[0116] In an embodiment of the present invention, the lower limit of the adjustable range of the second weight value is greater than or equal to the first weight value, thereby further improving the influence of operational information in the process of calculating the display value of existing commodities, making more obvious adjustments to the display position (or display order) of commodities with preset operational attributes, and further improving the effectiveness of the display.

[0117] In a specific implementation of step S16, at least a portion of the existing commodities are sorted and displayed based on the display value.

[0118] Specifically, all of the existing commodities may be sorted and displayed, or the improved display method in the embodiment of the present invention may be adopted to arrange them from large to small based on the display value, and then display a part of them.

[0119] Furthermore, after displaying a preset number of existing commodities, associated words may be used to determine and display related but not completely identical commodities with preset operation information.

[0120] Among them, associative words are used to indicate extended words that are related to the original words but not completely identical.

[0121] As a non-limiting example, the search information is "peanut oil". After displaying a preset number (such as 50) of "peanut oil" products, the associated word "soybean oil" can be used to search for existing products, and display self-operated products or "soybean oil" products from preset suppliers.

[0122] In an embodiment of the present invention, by segmenting the search information and using the search word set to determine the matching degree of each existing product, the products that match the search information input by the user can be displayed first. On this basis, by determining the operating information of each existing product, a weighted operation is performed at least based on the matching degree and the operating information to obtain the display value of the existing products one by one, and at least a part of the existing products are sorted and displayed. This can increase the dimension of analysis, consider more dimensions on the basis of the search information input by the user and the operating information of the product, and effectively improve the comprehensiveness and systematicness of the display. Furthermore, by using an adjustable second weight value as the weight of the operating information of each existing product, the corresponding weight can be flexibly configured according to the diversity of operating activities, such as priority display or delayed display, thereby further improving the effectiveness and accuracy of the display.

[0123] In a specific implementation of an embodiment of the present invention, the step of performing weighted operations at least based on the matching degree and operation information to obtain the display values ​​of existing commodities one by one may include: performing weighted operations based on the matching degree, operation information and a first parameter group to obtain the display values ​​of existing commodities one by one; wherein the first parameter group has a preset third weight value, and the third weight value is less than the first weight value, and less than the second weight value.

[0124] It can be understood that the display importance of the matching degree and operation information is higher than the display importance of the parameters in the first parameter group.

[0125] In an embodiment of the present invention, the weighted operation is performed at least based on the matching degree and the operation information to obtain the display value of the existing goods one by one, including: performing a weighted operation based on the matching degree, the operation information and the first parameter group to obtain the display value of the existing goods one by one; wherein the first parameter group has a smaller preset third weight value. The above scheme can be combined with the first parameter group to further increase the dimension of analysis, while ensuring the importance of the display of matching degree and operation information, by considering more dimensions, the comprehensiveness and systematicness of the display can be effectively improved.

[0126] Furthermore, the first parameter group includes one or more of the following: a level value belonging to the number of historical searches for each existing product; a level value belonging to the shelf time of each existing product; wherein, the more times the existing product is searched, the greater the level value belonging to the number of historical searches; and the shorter the shelf time of the existing product, the greater the level value belonging to the shelf time.

[0127] Referring to Table 1, Table 1 is an illustrative example table of a weighted logic in an embodiment of the present invention.

[0128] Table 1

[0129]

[0130]

[0131] For more information about the matching degree, please refer to the previous text and the matching degree determination formula, which will not be repeated here.

[0132] It should be pointed out that the historical search times and listing time can be divided into more levels and more detailed categories to obtain more detailed scores for subsequent weighted calculations, and are not limited to the two levels shown in Table 1.

[0133] Among them, the existing product has a preset operational attribute, and the promotion value is a third preset value. The existing product does not have a preset operational attribute, and the promotion value is a fourth preset value. The third preset value is greater than the fourth preset value. In Table 1, 100 is used to represent the third preset value, and -100 is used to represent the fourth preset value. In a specific implementation manner, the value selection method of the third preset value and the fourth preset value is not limited to this.

[0134] Furthermore, numerical values ​​with the same absolute value but different signs are used as scores of the operation information.

[0135] In the embodiment of the present invention, by using numerical values ​​with the same absolute value but different signs as the scores of the operation information, the display importance of the operation information can be further improved, and the differences in various possible scores of the operation information can be more significantly distinguished.

[0136] As shown in Table 1, the lower limit of the adjustable range of the second weight value of the operation information (such as 1000) can be greater than or equal to the first weight value of the matching degree (such as 500), and the third weight value of the first parameter group including the number of historical searches and the shelf time (such as 300) can be less than the first weight value and the second weight value to ensure the matching degree and the importance of displaying the operation information.

[0137] Among them, the full score can be used to limit the matching degree, operation information, and the upper limit of the score of each parameter in the first parameter group. The score of each parameter is weighted with the corresponding weight value, such as weighted summation, weighted average, and other appropriate weighted calculation methods.

[0138] It should be noted that when each weight index contains two or more parameters, a comprehensive score can be determined based on the scores of each parameter, and then a weighted operation can be performed. The step of determining the comprehensive score based on the scores of each parameter can be selected from: cumulative addition operation, cumulative average operation, weighted summation operation, weighted average operation, and other appropriate operation methods.

[0139] In another specific implementation of the embodiment of the present invention, the step of performing weighted operations at least based on the matching degree and operation information to obtain the display values ​​of existing commodities one by one may include: performing weighted operations based on the matching degree, operation information, a first parameter group and a second parameter group to obtain the display values ​​of existing commodities one by one; wherein the second parameter group has a preset fourth weight value, and the fourth weight value is less than the third weight value.

[0140] It can be understood that the display importance of the matching degree and operation information is higher than the display importance of the parameters in the first parameter group, and the display importance of the parameters in the first parameter group is higher than the display importance of the parameters in the second parameter group.

[0141] In an embodiment of the present invention, the weighted operation is performed at least based on the matching degree and the operation information to obtain the display value of the existing goods one by one, including: weighted operation is performed based on the matching degree, the operation information, the first parameter group and the second parameter group to obtain the display value of the existing goods one by one; wherein the second parameter group has a preset fourth weight value, and the fourth weight value is less than the third weight value. The above scheme can be combined with the second parameter group to further increase the dimension of analysis, while ensuring the matching degree, the operation information, and the display importance of the first parameter group, by considering more dimensions, the comprehensiveness and systematicness of the display can be effectively improved.

[0142] Furthermore, the second parameter group includes one or more of the following: a level value to which the historical click rate of each existing product belongs; a level value to which the historical sales volume of each existing product belongs; a promotion value of each existing product; wherein, the more times the existing product is clicked, the greater the level value to which the historical click rate belongs; the more historical sales volume of the existing product, the greater the level value to which the historical sales volume belongs; if the existing product has promotion information, the promotion value is a first preset value; if the existing product does not have promotion information, the promotion value is a second preset value, and the first preset value is greater than the second preset value.

[0143] Referring to Table 2, Table 1 is an example table illustrating another weighting logic in an embodiment of the present invention.

[0144] Table 2

[0145]

[0146]

[0147] For more information about matching degree, operation information, historical search times, and listing time, please refer to the previous text and the description in Table 2, which will not be repeated here.

[0148] It should be pointed out that the historical click rate and historical sales volume can be divided into more levels and more detailed categories to obtain more detailed scores for subsequent weighted calculations, and are not limited to the two levels shown in Table 2.

[0149] Among them, the existing product has preset promotion information, and the promotion value is a first preset value. The existing product does not have preset promotion information, and the promotion value is a second preset value. The first preset value is greater than the second preset value. In Table 2, 100 is used to represent the first preset value, and -100 is used to represent the second preset value. In a specific implementation manner, the value selection method of the first preset value and the second preset value is not limited to this.

[0150] Furthermore, numerical values ​​with the same absolute value but different signs are used as the score of the promotion information (ie, the promotion value).

[0151] In an embodiment of the present invention, by using numerical values ​​with the same absolute value but different signs as the score of the promotional information, the differences in various possible scores of the operational information can be more significantly distinguished, which helps to increase or decrease the display importance of the promotional information according to the specific circumstances.

[0152] Among them, the full score can be used to limit the upper limit of the score of each parameter in the matching degree, operation information, first parameter group, and second parameter group. The score of each parameter is weighted with the corresponding weight value, such as weighted summation, weighted average, and other appropriate weighted calculation methods.

[0153] It should be noted that when each weight index contains two or more parameters, a comprehensive score can be determined based on the scores of each parameter, and then a weighted operation can be performed. The step of determining the comprehensive score based on the scores of each parameter can be selected from: cumulative addition operation, cumulative average operation, weighted summation operation, weighted average operation, and other appropriate operation methods.

[0154] Furthermore, the first parameter group and / or the second parameter group can be divided into subgroups, each subgroup contains one or more parameters, and the preset weight values ​​between different subgroups are the same or different, thereby further improving the detailedness of the division of parameters and corresponding weight values.

[0155] Reference Figure 3 , Figure 3 : is a schematic diagram of the structure of a commodity display device in an embodiment of the present invention. The commodity display device may include:

[0156] An acquisition module 31, used to acquire search information;

[0157] A word segmentation module 32, used to segment the search information to determine a search word set;

[0158] A matching degree determination module 33, used to determine the matching degree of each existing product using the search word set;

[0159] An operation determination module 34 is used to determine the operation information of each existing commodity, wherein the operation information is at least used to indicate whether the existing commodity participates in any operation activities;

[0160] A weighting module 35, configured to perform a weighted operation based at least on the matching degree and the operation information to obtain display values ​​of the existing products one by one, wherein a preset first weight value is used as the weight of the matching degree of each existing product, and an adjustable second weight value is used as the weight of the operation information of each existing product;

[0161] The display module 36 is used to sort and display at least part of the existing commodities based on the display value.

[0162] about Figure 3 For more information about the working principle, working mode, beneficial effects, etc. of the commodity display device shown, please refer to the above text and related descriptions, which will not be repeated here.

[0163] The embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned commodity display method are executed. The storage medium may include ROM, RAM, a magnetic disk or an optical disk, etc. The storage medium may also include a non-volatile memory or a non-transitory memory, etc.

[0164] The embodiment of the present invention further provides a terminal, including a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor executes the steps of the above-mentioned commodity display method when running the computer program. The terminal includes but is not limited to terminal devices such as mobile phones, computers, tablet computers, servers, and cloud platforms.

[0165] It should be understood that in the embodiments of the present application, the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0166] It should also be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM) and direct rambus RAM (DR RAM).

[0167] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer program may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless means.

[0168] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may be physically included separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units. For example, for each device or product applied to or integrated in a chip, each module / unit contained therein may be implemented in the form of hardware such as circuits, or at least part of the modules / units may be implemented in the form of software programs, which run on the processor integrated inside the chip, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits; for each device or product applied to or integrated in a chip module, each module / unit contained therein may be implemented in the form of hardware such as circuits, and different modules / units may be located in the same component (such as a chip, a circuit module, etc.) or different components of the chip module, or at least part of the modules / units may be implemented in the form of software programs, which run on the processor integrated inside the chip module, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits.

[0169] It should be understood that the term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article indicates that the associated objects before and after are in an "or" relationship.

[0170] The "plurality" appearing in the embodiments of the present application refers to two or more.

[0171] The first, second, etc. descriptions appearing in the embodiments of the present application are only used for illustration and distinction of the description objects. There is no order, nor do they indicate any special limitation on the number of devices in the embodiments of the present application, and cannot constitute any limitation on the embodiments of the present application.

[0172] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the scope defined by the claims.

Claims

1. A commodity display method, characterized in that: include: Get search information; Segmenting the search information to determine a search word set; Determine the matching degree of each existing product by using the search word set; Determining the operation information of each existing commodity, wherein the operation information is at least used to indicate whether the existing commodity has a preset operation attribute; A weighted calculation is performed based at least on the matching degree and the operation information to obtain display values ​​of the existing products one by one, wherein a preset first weight value is used as the weight of the matching degree of each existing product, and an adjustable second weight value is used as the weight of the operation information of each existing product; At least a portion of the existing products are sorted and displayed based on the display value.

2. The commodity display method according to claim 1, characterized in that: The operational information is selected from one or more of the following: Whether each existing product is a self-operated product; Whether each existing product belongs to the preset supplier.

3. The commodity display method according to claim 1 or 2, characterized in that: The lower limit of the adjustable range of the second weight value is greater than or equal to the first weight value.

4. The commodity display method according to claim 1, characterized in that: The search information is segmented to determine a search word set, including: Preprocessing the search information to obtain a preprocessing result; Using a reverse maximum matching algorithm, each word in the preprocessing result is matched in a word database until a matching word is found for each word or only a single word remains; The search word set is constructed, and the search word set includes each matching word and its replacement word.

5. The commodity display method according to claim 4, characterized in that: Preprocessing the search information to obtain a preprocessing result includes: Filtering the search information with preset words, wherein the preset words include one or more of the following: modal particles, auxiliary words, garbled characters, sensitive words, and stop words; The filtered search information is segmented to obtain the preprocessing result.

6. The commodity display method according to claim 4, characterized in that: The replacement word for each matching word is selected from: a synonym of the matching word and / or a near synonym of the matching word.

7. The commodity display method according to claim 1, characterized in that: The matching degree of each existing product is determined by using the search word set, including: According to the data vector of the preset dimension, the search word set is used to construct the data vector to be searched, wherein the data of at least a part of the positions in the data vector of the preset dimension is used to indicate at least a part of the words in the word database in a one-to-one correspondence; The cosine similarity algorithm is used to compare the data vector to be searched with the data vector of each existing product one by one to determine the matching degree of each existing product.

8. The commodity display method according to claim 7, characterized in that: According to the data vector of the preset dimension, the search word set is used to construct the data vector to be searched, including: The data vector of the preset dimension is traversed, and the positions of the words existing in the search word set are assigned with the first preset value, and the remaining positions are assigned with the second preset value, so as to obtain the data vector to be searched.

9. The commodity display method according to claim 7, characterized in that: Before comparing the to-be-searched data vector with the data vector of each existing product one by one, the method further includes: Segment the product information of each existing product to obtain a product word set and a word database for each existing product; For each existing product, the data vector of the preset dimension is traversed, the position of the word existing in the product word set is assigned a first preset value, and the remaining positions are assigned a second preset value to obtain the data vector of each existing product.

10. The commodity display method according to any one of claims 7 to 9, characterized in that: The data vector of the preset dimension is a single-row and N-column vector or an N-row and single-column vector; Wherein, N is used to represent the preset dimension, and N is greater than or equal to the number of words in the word database.

11. The commodity display method according to claim 1, characterized in that: The performing weighted calculation based at least on the matching degree and the operation information to obtain the display value of the existing products one by one includes: Performing a weighted calculation based on the matching degree, the operation information and the first parameter group to obtain display values ​​of the existing products one by one; The first parameter group has a preset third weight value, and the third weight value is smaller than the first weight value and smaller than the second weight value.

12. The commodity display method according to claim 11, characterized in that: The first parameter group includes one or more of the following: The level value of each existing product's historical search count; The level value to which each existing product's listing time belongs; The more times the existing product is searched, the greater the level value of the historical search times; The shorter the shelf time of the existing product, the greater the grade value to which the shelf time belongs.

13. The commodity display method according to claim 11, characterized in that: The performing weighted calculation based at least on the matching degree and the operation information to obtain the display value of the existing products one by one includes: Performing a weighted calculation based on the matching degree, the operation information, the first parameter group, and the second parameter group to obtain display values ​​of the existing products one by one; The second parameter group has a preset fourth weight value, and the fourth weight value is smaller than the third weight value.

14. The commodity display method according to claim 13, characterized in that: The second parameter group includes one or more of the following: The level value to which the historical click rate of each existing product belongs; The level value to which each existing product's historical sales volume belongs; The promotion value of each existing product; The more times the existing product is clicked, the greater the level value of the historical click rate; The more historical sales volume of the existing product, the greater the level value to which the historical sales volume belongs; The existing product has promotion information, and the promotion value is a first preset value. The existing product does not have promotion information, and the promotion value is a second preset value. The first preset value is greater than the second preset value.

15. A commodity display device, characterized in that: include: An acquisition module, used to acquire search information; A word segmentation module, used to segment the search information to determine a search word set; A matching degree determination module, used to determine the matching degree of each existing product using the search word set; An operation determination module, used to determine the operation information of each existing commodity, wherein the operation information is at least used to indicate whether the existing commodity participates in any operation activities; A weighting module, configured to perform a weighted operation based at least on the matching degree and the operation information to obtain display values ​​of the existing products one by one, wherein a preset first weight value is used as the weight of the matching degree of each existing product, and an adjustable second weight value is used as the weight of the operation information of each existing product; A display module is used to sort and display at least part of the existing commodities based on the display value.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the commodity display method according to any one of claims 1 to 14 are executed.

17. A terminal comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, characterized in that: When the processor runs the computer program, the steps of the commodity display method according to any one of claims 1 to 14 are performed.

Images