CN110580649B

CN110580649B - Method and device for determining commodity potential value

Info

Publication number: CN110580649B
Application number: CN201810586801.9A
Authority: CN
Inventors: 余帅兵; 王泉泉
Original assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Jingdong Shangke Information Technology Co Ltd
Current assignee: Beijing Jingdong Century Trading Co Ltd; Beijing Jingdong Shangke Information Technology Co Ltd
Priority date: 2018-06-08
Filing date: 2018-06-08
Publication date: 2024-03-05
Anticipated expiration: 2038-06-08
Also published as: CN110580649A

Abstract

The invention discloses a method and a device for determining commodity potential values, and relates to the technical field of computers. One embodiment of the method comprises the following steps: acquiring commodities and determining attribute values of each commodity; based on a preset statistical model, counting the attribute values of each commodity to obtain an attribute value combination; based on the attribute value of each commodity, determining the commodity corresponding to the attribute value combination, and determining the potential value of the attribute value combination according to the category number of the commodity corresponding to the attribute value combination and sales information in a preset history time. The embodiment can effectively understand the preference and consumption behavior habit of the consumer, further effectively analyze the combination characteristics of the commodity attribute values, and provide basis for mining and identifying future market segments with sales growth potential by combining big data and learning methods thereof.

Description

Method and device for determining commodity potential value

Technical Field

The present invention relates to the field of computer technologies, and in particular, to a method and an apparatus for determining a commodity potential value.

Background

With the development of networks, the status of electronic commerce in the whole business field is gradually improved, and more users select online shopping modes. In order to meet various demands of users, an electronic commerce platform is expected to accurately grasp the change of demand trend of consumers on commodities. In addition, upstream users (e.g., manufacturers) as commodity production supply chains also want to know the market blue sea with potential in the future through various sales channels in order to make product production planning early.

At present, the market is usually provided with consultation and research modes, the preferences (such as price, materials and popular trends) of various consumers on certain types of commodities are collected, the characteristics of products which accord with the preference and tendency of the consumers are further improved, and market positioning is carried out according to research results and new products are developed in an adaptive mode.

As an alternative implementation manner of the above mode, for a large-scale electronic commerce platform, at present, a machine learning and big data mode is mainly adopted, and various commodity and consumer data stored in the platform are mined, clustered, analyzed and the like to find out the relevance and consumption rule between the commodity and the consumer so as to provide product development suggestions for upstream users.

In carrying out the present invention, the inventors have found that at least the following problems exist in the prior art:

1) The dividing dimension of the market is rough: at present, the industry is mainly taken as a division basis, so that granularity of a drilling and research market is rough, and the complex, rich and personalized requirements of consumers cannot be met finely;

2) The market potential is mined and analyzed without a systematic, movable and reusable method, the conclusion of market research and analysis is limited by the collected samples, a great deal of time and labor cost are required, the development period from the conclusion to the product is longer, and scientific and quantized identification of the market potential cannot be achieved.

Disclosure of Invention

In view of the above, the embodiment of the invention provides a method and a device for determining commodity potential values, which at least can solve the problem that the prior systematic method for mining and identifying commodity potential causes coarse dimension of market division and cannot meet the demands of various users.

To achieve the above object, according to an aspect of the embodiments of the present invention, there is provided a method for determining a commodity potential value, including:

acquiring commodities and determining attribute values of each commodity;

based on a preset statistical model, counting the attribute values of each commodity to obtain an attribute value combination;

based on the attribute value of each commodity, determining the commodity corresponding to the attribute value combination, and determining the potential value of the attribute value combination according to the category number of the commodity corresponding to the attribute value combination and sales information in a preset history time.

Optionally, acquiring the commodities, determining the attribute value of each commodity, and further including:

verifying the attribute value of each commodity based on a preset verification rule, and extracting commodities of which the attribute values do not accord with the verification rule; determining similar commodities associated with the extracted commodities, and acquiring attribute values of the similar commodities; when the attribute values of the similar commodities accord with the verification rule, replacing the attribute values of the extracted commodities to be the attribute values of the similar commodities.

Optionally, before verifying each attribute value based on a predetermined verification rule, the method further includes: and acquiring characteristic information of each commodity, obtaining the similarity among the commodities according to a preset similarity calculation mode, and determining the commodities with the similarity exceeding a preset similarity threshold as similar commodities.

determining the attribute to which each attribute value belongs and the category to which each attribute belongs, and dividing the attributes belonging to the same category into a group; acquiring commodity operation behavior information of a user under each attribute, combining a preset decision model to acquire the importance of each attribute, and extracting the attribute of which the importance exceeds a corresponding preset importance threshold under each category; and determining the attribute value associated with the extracted attribute of each commodity according to the attribute value of each commodity and the attribute to which each attribute value belongs.

Optionally, acquiring the commodities, and determining the attribute value of each commodity includes: acquiring commodities, determining commodity information of each commodity, and splitting each commodity information based on a preset splitting mode to obtain corresponding attribute values.

Optionally, based on a predetermined statistical model, counting attribute values of each commodity to obtain an attribute value combination, including: based on a preset statistical model, counting attribute values of each commodity, and determining the occurrence number of each attribute value combination; and determining the support degree of each attribute value combination according to the occurrence times of each attribute value combination and the number of all commodities, and extracting the attribute value combination with the support degree exceeding a preset support degree threshold.

Optionally, determining the potential value of the attribute value combination according to the number of kinds of commodities corresponding to the attribute value combination and sales information in a predetermined history time length includes: acquiring the occurrence times of each attribute value combination, determining the frequency of each attribute value combination by combining sales information of commodities corresponding to each attribute value combination in a preset historical time, extracting attribute value combinations with the frequency exceeding a preset frequency threshold, and determining the extracted attribute value combinations as first attribute value combinations; acquiring the category number and market capacity of the commodities corresponding to each first attribute value combination, extracting first attribute value combinations with the category number exceeding a preset category threshold value and the market capacity not exceeding the preset market capacity threshold value, and determining the extracted first attribute value combinations as second attribute value combinations; and determining corresponding potential values of each second attribute value combination according to the category number of the commodity corresponding to each second attribute value combination and sales information in the preset historical time.

To achieve the above object, according to another aspect of the embodiments of the present invention, there is provided a device for determining a commodity potential value, including:

The attribute value acquisition module is used for acquiring commodities and determining the attribute value of each commodity;

the attribute value combination module is used for counting the attribute value of each commodity based on a preset statistical model to obtain attribute value combination;

the potential value determining module is used for determining commodities corresponding to the attribute value combinations based on the attribute values of each commodity, and determining potential values of the attribute value combinations according to the types and the numbers of the commodities corresponding to the attribute value combinations and sales information in a preset historical time.

Optionally, the attribute value acquisition module is further configured to: verifying the attribute value of each commodity based on a preset verification rule, and extracting commodities of which the attribute values do not accord with the verification rule; determining similar commodities associated with the extracted commodities, and acquiring attribute values of the similar commodities; and when the attribute values of the similar commodities accord with the verification rule, replacing the attribute values of the extracted commodities to be the attribute values of the similar commodities.

Optionally, the attribute value acquisition module is further configured to: and acquiring characteristic information of each commodity, obtaining the similarity among the commodities according to a preset similarity calculation mode, and determining the commodities with the similarity exceeding a preset similarity threshold as similar commodities.

Optionally, the attribute value acquisition module is further configured to:

Optionally, the attribute value acquisition module is configured to: acquiring commodities, determining commodity information of each commodity, and splitting each commodity information based on a preset splitting mode to obtain corresponding attribute values.

Optionally, the attribute value combination module is configured to: based on a preset statistical model, counting attribute values of each commodity, and determining the occurrence number of each attribute value combination; and determining the support degree of each attribute value combination according to the occurrence times of each attribute value combination and the number of all commodities, and extracting the attribute value combination with the support degree exceeding a preset support degree threshold.

Optionally, the potential value determining module is configured to: acquiring the occurrence times of each attribute value combination, determining the frequency of each attribute value combination by combining sales information of commodities corresponding to each attribute value combination in a preset historical time, extracting attribute value combinations with the frequency exceeding a preset frequency threshold, and determining the extracted attribute value combinations as first attribute value combinations; acquiring the category number and market capacity of the commodities corresponding to each first attribute value combination, extracting first attribute value combinations with the category number exceeding a preset category threshold value and the market capacity not exceeding the preset market capacity threshold value, and determining the extracted first attribute value combinations as second attribute value combinations; and determining corresponding potential values of each second attribute value combination according to the category number of the commodity corresponding to each second attribute value combination and sales information in the preset historical time.

To achieve the above object, according to still another aspect of the embodiments of the present invention, there is provided an electronic device for determining a commodity potential value.

The electronic equipment of the embodiment of the invention comprises: one or more processors; and the storage device is used for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors are enabled to realize the method for determining the commodity potential value.

To achieve the above object, according to still another aspect of the embodiments of the present invention, there is provided a computer readable medium having stored thereon a computer program which, when executed by a processor, implements the method of determining a commodity potential value according to any one of the above.

According to the solution provided by the present invention, one embodiment of the above invention has the following advantages or beneficial effects: the method can effectively learn the preference and the consumption behavior habit of the consumer, further effectively analyze the combination characteristics of the commodity attribute values, and mine and identify the market segment with sales growth potential in the future for the user by combining big data and learning methods thereof.

Further effects of the above-described non-conventional alternatives are described below in connection with the embodiments.

Drawings

The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:

FIG. 1 is a schematic flow chart of a method for determining a commodity potential value according to an embodiment of the present invention;

FIG. 2 is a flow chart of an alternative method of determining commodity potential values according to an embodiment of the present invention;

FIG. 3 is a flow chart of another alternative method of determining commodity potential values according to an embodiment of the present invention;

FIG. 4 is a flow chart of yet another alternative method of determining commodity potential values according to an embodiment of the present invention;

FIG. 5 is a flow chart of yet another alternative method of determining commodity potential values in accordance with an embodiment of the present invention;

FIG. 6 is a flow chart of yet another alternative method of determining commodity potential values according to an embodiment of the present invention;

FIG. 7 is a flow chart of yet another alternative method of determining commodity potential values according to an embodiment of the present invention;

FIG. 8 is a flow chart of a method for determining a potential value of a particular commodity according to an embodiment of the present invention;

FIG. 9 is a schematic diagram of the main modules of a commodity potential value determining apparatus according to an embodiment of the present invention;

FIG. 10 is an exemplary system architecture diagram in which embodiments of the present invention may be applied;

Fig. 11 is a schematic diagram of a computer system suitable for use in implementing the mobile device or server of an embodiment of the present invention.

Detailed Description

Exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding, and are to be considered merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

It should be noted that the present invention focuses only on how to mine and identify the potential of the sub-divided commodity and does not directly locate the exploded commodity in the market.

The embodiment provided by the invention can be applied to various industries and categories, such as consumer product industry, household appliance industry, electronic data industry and the like, and has strong migration and reusability.

A "market segment" is generally understood to be defined as: according to different consumer groups, the market is divided into a plurality of sub-market which can meet the demands of different consumers. It comprises three layers: the first layer is for only one commodity class, the second layer is for one commodity attribute value under one class, and the third layer is for a combination of commodity attribute values.

The market segment expressed by the invention is equivalent to attribute value combination in the scene of potential market. Users (e.g., brands) may host different consumer markets or consumer groups by producing different types of goods.

The terms related to the present invention are explained as follows:

market blue sea: representing a market segment of goods that are not yet perceived but have a good growth potential.

All kinds of consumers: consumer groups of different ages, sexes, territories, consumption capacities, industries.

Market positioning: refers to an arrangement that is made to take a clear, specific and ideal position of the product in the mind of the intended consumer relative to the competing product.

Three classes: the primary and secondary categories are summarized, and the tertiary is a very detailed category; such as clothing footwear caps: primary category refers to: clothing underwear, second category refers to: men's wear, women's wear, underwear, tertiary categories refer to: scarf/glove/hat sets, etc., the present invention is described in three levels of categories only.

Frequent collection: there are a series of sets with identical elements, and elements with high frequency of simultaneous occurrence in the sets form a subset, meeting a certain threshold condition.

Frequent pattern: frequently occurring item sets, sequences or substructures in a dataset; for example, in shopping basket analysis, it is analyzed which items are frequently purchased simultaneously by the user.

Support degree: the probability that some two (several) items are purchased simultaneously (here, simultaneously, generally meaning with a single or one independent transaction) in all transactions analyzed is abbreviated as the percentage of simultaneous inclusion X, Y in the transaction.

Consumer decision tree: one of the algorithmic modeling of machine learning simulates a consumer's delicately evaluating the attributes of a product, brand, or service, and performs the process of selecting, purchasing, and ordering a product that meets a particular need to determine the consumer's importance and order of product attributes.

FP-growth model: one of the algorithmic modeling of machine learning can be used to perform data mining, computing relevance and frequent features.

Referring to fig. 1, a main flowchart of a method for determining a commodity potential value according to an embodiment of the present invention is shown, including the following steps:

s101: acquiring commodities and determining attribute values of each commodity.

S102: and counting the attribute values of each commodity based on a preset statistical model to obtain an attribute value combination.

S103: based on the attribute value of each commodity, determining the commodity corresponding to the attribute value combination, and determining the potential value of the attribute value combination according to the category number of the commodity corresponding to the attribute value combination and sales information in a preset history time.

In the above embodiment, in step S101, the commodity is generally provided with the commodity main database in the electronic commerce data background, and the attribute data of the commodity is recorded.

In the commodity main database, each commodity has independent information, and commodity attribute values or commodity attributes of the commodity are recorded, for example, biscuit-taste-chocolate taste, and the chocolate taste is the attribute value.

The method is mainly applied to the electronic commerce platform, and only the commodities under the three-level class are processed, so that the number of the obtained commodities is controllable. In addition, when the number of commodities is too large, distributed computing and other modes are supported.

For step S102, some of the products contain the same attribute values, such as shampoo 1-oil control, shampoo 2-oil control, and softening.

When the same attribute value appears in the attribute value sets of two commodities, the attribute value can be mapped to the two commodities, the attribute values which appear at the same time and have higher frequency can be combined to generate an attribute value combination in a frequent data statistics mode.

In addition, since the number of attribute values of the commodity is large, the number of generated attribute value combinations is large, and in order to reduce the subsequent calculation amount, a certain threshold condition needs to be satisfied for the combination of attribute values, for example, two identical attribute values, and only when the number of simultaneous occurrences exceeds a set number (for example, 5 times), the combination can be made into one attribute value combination.

The attribute value combination counted according to the frequent pattern is defined as a market segment under a three-level category, and the market segment is short for convenience of follow-up overview.

For the segment market obtained in step S103, there are combinations among attribute values of a single attribute, for example, shampoo-efficacy-moisturizing+nourishing, and there are N pairs of paired combination relationships of a single attribute and other attributes of two, three, four, and four, for example, shampoo-efficacy+capacity-moisturizing and nourishing of >300ml, and the final attribute value combinations may be up to thousands, i.e. thousands of segment markets can be segmented under one three-level category. Therefore, there is a need to find a batch from these thousands of market segments that has a better sales growth potential.

For analysis of market segments potential, commodities meeting the attribute value combination condition can be extracted according to commodity attribute value combinations corresponding to each market segment place to form a commodity set.

It should be noted that, there may be a plurality of attribute values of a commodity, and the attribute value combination only includes some attribute values, so some commodities may correspond to a plurality of attribute value combinations, that is, the commodity may have a commodity set of a plurality of attribute value combinations at the same time, and both the commodity and the attribute value combination may be in a mapping relationship of many to many.

The potential value for each market segment can be determined from the sales information and the market capacity (i.e., the number of available categories of goods) for each set of goods. Specifically, the contribution amount, i.e., the potential value, of each segment market in the entire market is determined by both.

In which it is to be noted that,

1) The market capacity is the number of available goods in the market, which is distinguished from the stock quantity of goods, representing the number of goods types. For example, there are 20 inventory items indicating that there are 20 items to sell, but the 20 items may be just one SKU, i.e., one valid item. Market capacity of the market segment can be obtained by counting SKU types of all commodities in the corresponding commodity set.

2) The sales information may include one of sales or sales;

(1) the sales volume is effective commodity sales volume, namely data of commodity sales volume is actually generated, because the situation that an order is generated but payment is not completed finally exists; when sales were 1000 and market capacity was 5, the resulting potential value was 1000/5=200;

(2) for sales, the average price of sales in a predetermined history period can be combined on the basis of acquiring sales, that is, sales amount= sales amount x average sales price of commodity. The average price may be determined by analyzing price fluctuations of the commodity over a historical period of time. The total sales of the market segment can be obtained by counting sales of each commodity in the corresponding commodity set; the resulting potential value was 2000 when sales were 1000, historical average price was 10, market capacity was 5.

Further, if the obtained potential values are all larger, the potential values may be processed for subsequent calculation, for example, each potential value is divided by 10000, and the potential value is 2000→0.2, and the processing method is not limited herein.

Further, for the obtained potential contribution values, the final ranking of the potential market is obtained according to the order of the values from high to low. For example, only the attribute value combination with the highest ranking first potential is extracted, or the attribute value combination of the first three of the ranking is extracted as the main research direction of the subsequent market.

The method provided by the embodiment provides a market blue sea identification model, and the potential value of each market segment is determined based on the market capacity and commodity sales information of the market segment, so that blue sea areas with sales growth potential are located from thousands of market segments. The method greatly reduces the cost of the traditional market research, improves the commodity division fineness based on the attribute value combination mode, and provides a solid foundation for the subsequent market decision.

Referring to fig. 2, a flowchart of an alternative method for determining a commodity potential value according to an embodiment of the present invention is shown, including the following steps:

S201: acquiring commodities and determining attribute values of each commodity.

S202: and verifying the attribute value of each commodity based on a preset verification rule, and extracting the commodity with the attribute value not conforming to the verification rule.

S203: and determining similar commodities associated with the extracted commodities, and acquiring attribute values of the similar commodities.

S204: when the attribute values of the similar commodities accord with the verification rule, replacing the attribute values of the extracted commodities to be the attribute values of the similar commodities.

S205: and counting the attribute values of each commodity based on a preset statistical model to obtain an attribute value combination.

S206: and determining attribute value combinations to which each commodity belongs based on the attribute values of each commodity, and determining potential values of the attribute value combinations according to the types and the numbers of the commodities corresponding to the attribute value combinations and sales information in a preset historical time.

In the above embodiment, steps S201, S205 and S206 can be referred to the descriptions of steps S101, S102 and S103 shown in fig. 1, and are not repeated here.

In the bottom layer data, the maintenance of commodity attribute data may be wrong or missing due to historical system reasons, so that before the attribute value combination is generated by combining attribute values, the attribute values of the commodity can be checked and filled first, so that the integrity and effectiveness of the attribute data are ensured, and the accuracy of subsequent data processing is improved.

In the above embodiment, the verification of the attribute values in steps S202 and S203 is mainly performed by machine learning similarity+artificial rule setting.

The similarity of the commodities may be determined based on feature information such as an outline, a picture, and a description of each commodity. For example, both the commodity 1 and the commodity 2 are shipped from the same brand a, and the description information of both the commodities is basically consistent, that is, the commodities are determined to be similar.

Note that, for the commodity and the related similar commodity, both belong to the commodity in step S201.

For checking the commodity attribute value, it is mainly determined whether the attribute value is complete, i.e. null. For the condition that the commodity characteristic information is incomplete, in order to reduce the error rate of the subsequent attribute value combination, the attribute value needs to be filled in a certain mode.

The specific process comprises the following steps:

1) If the attribute value is not null, not performing any processing;

2) If the attribute value is null, the attribute value of the commodity is incomplete, and the commodity needs to be filled:

(1) if the similar commodity is associated with the commodity and the attribute value of the similar commodity is complete, the attribute value of the similar commodity is given to the attribute value of the incomplete commodity, namely technical filling is carried out;

It should be noted that, for the replacement of the attribute value, the attribute information of the commodity which is relatively perfect in the similar commodity is filled into the commodity which is relatively imperfect in the commodity attribute, and only the attribute value which is originally empty is replaced, and for the attribute values which are complete in other parts, the replacement is not needed.

(2) There is a similar commodity associated with the commodity, but the attribute value of the similar commodity is also incomplete, i.e. the attribute value of the commodity is a part of empty, the similar commodity is also empty, and no processing is performed for the case.

(3) There is no similar merchandise associated with the merchandise and no processing is done for this case as well.

And finally, manually checking and taking care of the attribute values obtained by machine verification.

Further, there may be a certain execution order for the verification filling of the attribute values. In general, commodity attributes can be divided into "basic attributes", "function expansion attributes", "custom attributes such as specification and package" from coarse to fine. Wherein the first two attributes are relatively regular templates, and fields and definitions are relatively uniform; the last attribute is maintained by daily manual setting of merchants and operators, fields and definitions are relatively non-standard, and the last attribute is considered only when the current two attributes cannot meet business requirements, and is not considered in the usual case.

The related commodity attribute values and commodity attributes have a mapping relation, and the commodity attribute values and commodity attributes are stored in a field table in a database, for example, the commodity attributes are color (attr_name=color), and a plurality of corresponding commodity attribute values are provided (attr_value=yellow; red; green … …).

Referring to table 1, a comparison of the values of the attributes before and after repair is shown using the basic attributes as an example. The null value is displayed as an empty value of the attribute_name, or is not filled in an initial state, or is not recognized by filling in errors. As can be seen from table 1, the integrity of the bottom layer of the commodity attribute in the original data is poor, the empty value is relatively high, and after the technical processing filling, the null value in the table is recorded by the original 8703 pieces and is reduced to 2500 pieces.

TABLE 1 daily Property-Property value Table before repair → after repair

According to the method provided by the embodiment, before the attribute values are combined, the attribute values are checked and filled, so that the integrity and effectiveness of the follow-up data can be ensured, and the accuracy of market potential analysis is further improved. In addition, the provided method can process the attribute values of the first layer and the second layer, and compared with the traditional one-layer attribute maintenance, the method is higher in refinement degree.

Referring to fig. 3, a main flow chart of another alternative method for determining a commodity potential value according to an embodiment of the present invention is shown, including the following steps:

S301: acquiring commodities and determining attribute values of each commodity.

S302: the attribute to which each attribute value belongs and the category to which each attribute belongs are determined, and the attributes belonging to the same category are divided into a group.

S303: and acquiring commodity operation behavior information of the user under each attribute, and combining a predetermined decision model to obtain the importance of each attribute.

S304: and extracting the attribute of which the importance degree under each category exceeds a corresponding preset importance degree threshold value.

S305: and determining the attribute value associated with the extracted attribute of each commodity according to the attribute value of each commodity and the attribute to which each attribute value belongs.

S306: and counting the attribute values of each commodity based on a preset statistical model to obtain an attribute value combination.

S307: based on the attribute value of each commodity, determining the commodity corresponding to the attribute value combination, and determining the potential value of the attribute value combination according to the category number of the commodity corresponding to the attribute value combination and sales information in a preset history time.

In the above embodiment, step S301 may be described with reference to step S101 shown in fig. 1, or may be described with reference to steps S201 to S204 shown in fig. 2; the steps S306 and S307 can be referred to the descriptions of the steps S102 and S103 shown in fig. 1, and are not repeated here.

In the above embodiment, for step S302, some attributes have little influence on the market potential analysis, and these attributes and the attribute values included therein can be eliminated.

For the screening of attribute values, the screening needs to be performed according to the attribute, and the commodity attribute is only meaningful under the same category. For example, the commodity attributes of a beverage, etc., are taste, place of origin, brand, etc., but may not be applicable to another category of food.

For importance ranking of attributes, one can start with a minimum level category analysis, e.g., a three-level category, to analyze the market segment corresponding to different categories and determine the important attributes and their ranking that are most of interest to the consumer when purchasing the market segment's merchandise.

For step S303, according to the consumer decision tree model, the operational behavior information of the user on the same category of commodity under the e-commerce platform, for example, operational behavior information within one year, is analyzed.

The commodity may be the commodity obtained in step S301. The operation behavior information may be a history order record, a history browse record, etc. of the user on the e-commerce platform, where, for the user, only the user under one category is recorded.

Based on the commodity attributes, the acquired operation behavior information is counted and classified to calculate the jump and selection probability of the user between different attribute values. And ordering the attributes according to the obtained probability, so that the importance of each attribute can be obtained, the importance can be replaced by an attribute index, and the attribute index can be approximately understood as the index expression of the jump and selection probability.

Taking biscuits as an example, the attributes of the biscuits include taste types, places of production, types and the like, and the importance of each attribute is shown in table 2:

table 2 product attribute priority

For the priority of a product attribute, it may also represent a substitution between different attributes, the lower the substitution level, the higher the "importance" level of that attribute.

Additionally, the attributes may also be filtered according to the personalized needs of the user (e.g., vendor). For example, the supplier may produce biscuits with only one taste, but the categories may be varied, and the attribute of taste category may not be taken into account.

According to the method provided by the embodiment, the attribute with lower influence on the market potential analysis and the attribute value contained in the attribute can be removed through analyzing the importance of the attribute, so that the subsequent calculated amount is reduced, and the market potential analysis efficiency is improved.

Referring to fig. 4, a main flowchart of a method for determining a commodity potential value according to another embodiment of the present invention is shown, including the following steps:

s401: acquiring commodities, determining commodity information of each commodity, and splitting each commodity information based on a preset splitting mode to obtain corresponding attribute values.

S402: and counting the attribute values of each commodity based on a preset statistical model to obtain an attribute value combination.

S403: based on the attribute value of each commodity, determining the commodity corresponding to the attribute value combination, and determining the potential value of the attribute value combination according to the category number of the commodity corresponding to the attribute value combination and sales information in a preset history time.

In the above embodiment, for the steps S402 and S403, reference may be made to the descriptions of the steps S102 and S103 shown in fig. 1, and the descriptions are not repeated here.

In the database, there may be cases where there is insufficient data in the attribute values of the stored commodity, and some attribute values still exist in the commodity information thereof, for example, description information, commodity introduction, and the like. Therefore, to improve the integrity of subsequent attribute values, the attribute values of the commodity may be mined based on commodity information prior to combining the attribute values.

In the above embodiment, for step S401, the commodity information is a comprehensive description of the commodity, and for the commodity attribute value, the commodity information may be acquired based on the splitting of the commodity information.

For example, the main data table of the commodity can be subjected to data mining through an FP-Growth Model (frequent pattern discovery Model) machine learning method, so as to represent commodity information as a set of discretized vector values, and the vector values are split to obtain qualitative description of the commodity, and the qualitative description is encoded as a combination of commodity attribute values.

For example, the commodity information of one commodity is described as "20-inch hanging type liquid crystal black shell ultrathin television", and for this description, it can be quantized into a set of attribute combinations of four attributes (size+hanging mode+color+television type) of a\b\c\d, and further encoded into a similar 1001+2002+3001+4005 for machine identification.

The split involved is to express an attribute (e.g., color) as a set of attribute values, which facilitates subsequent combinations of commodity attribute values.

Further, the attribute according to which the attribute value is obtained by splitting the commodity information may be the attribute obtained by screening in fig. 3, where the obtained attribute value has pertinence, for example, the attribute obtained by screening is a biscuit taste, and the attribute value of the original commodity has no taste, but the commodity description information has a spicy taste, that is, the spicy taste is added to the attribute value of the commodity.

The method provided by the embodiment provides a way for acquiring the commodity attribute value, reduces the missing acquisition condition of the attribute value, and further improves the coverage range of the subsequent attribute value combination. In addition, the accuracy of the calculation result of the subsequent potential analysis is improved for the screening of the attribute values.

Referring to fig. 5, a main flowchart of a method for determining a commodity potential value according to an embodiment of the present invention is shown, including the following steps:

s501: acquiring commodities and determining attribute values of each commodity.

S502: and counting the attribute values of each commodity based on a predetermined statistical model, and determining the occurrence number of each attribute value combination.

S503: and determining the support degree of each attribute value combination according to the occurrence times of each attribute value combination and the number of all commodities.

S504: and extracting attribute value combinations with support exceeding a predetermined support threshold.

S505: based on the attribute value of each commodity, determining the commodity corresponding to the attribute value combination, and determining the potential value of the attribute value combination according to the category number of the commodity corresponding to the attribute value combination and sales information in a preset history time.

In the above embodiment, step S501 may refer to the description of step S101 shown in fig. 1, the descriptions of steps S201 to S204 shown in fig. 2, the descriptions of steps S301 to S304 shown in fig. 3, and the descriptions of step S401 shown in fig. 4; step S505 can be described with reference to step S103 shown in fig. 1, and will not be described herein.

The support degree provided refers to parameters and methods that ensure sample rationality while keeping data size within a certain controllable range in order to achieve a usable data set. For data sets with relevance, screening and acquisition are typically performed using a support-degree approach.

In the above embodiment, for step S502, for the market segment, i.e., the attribute value combination, the number of occurrences of different attribute values in the same product may be determined based on the machine learning method, for example, FP-Growth Model, by performing frequent-set data statistics on all products and their attribute values.

For step S503, for the support degree between different attribute values in the attribute value combination, there may be the number of times that different attribute values occur simultaneously in the attribute value of each commodity/the number of all commodities.

For step S504, to reduce the calculation amount of the subsequent market segment and ensure the calculation accuracy, only part of the attribute values may be combined, and the screening process is performed according to the minimum support.

The minimum support is the minimum threshold that is met by a data calculation in the FP-Growth Model to find the frequent item set available. Frequent sets may generate tens of thousands of pairs of relationships, the lower the minimum support, the more attribute value combinations are formed, and conversely, the fewer attribute value combinations are formed.

For example, there are 5 shampoos, the number of times that moisturization and softness occur simultaneously is 4, and the support of these two attribute values is 4/5=80%; the moisturizing, softening and nourishing occur for 3 times simultaneously, and the support degree between the three attribute values is 3/5=60%. When the minimum support is 70%, only the combination of "moisturization and softness" is extracted.

According to the method provided by the embodiment, the support degree among the attribute values is fully considered, so that the attribute value combination is screened, and the subsequent potential analysis amount and the background service pressure are reduced.

Referring to fig. 6, a main flowchart of a method for determining a commodity potential value according to an embodiment of the present invention is shown, including the following steps:

s601: acquiring commodities and determining attribute values of each commodity.

S602: and counting the attribute values of each commodity based on a preset statistical model to obtain an attribute value combination.

S603: and acquiring the occurrence times of each attribute value combination, and determining the frequency of each attribute value combination by combining sales information of the commodity corresponding to each attribute value combination in a preset historical time.

S604: and extracting attribute value combinations with the frequency exceeding a preset frequency threshold value, and determining the extracted attribute value combinations as first attribute value combinations.

S605: and obtaining the category number and the market capacity of the commodity corresponding to each first attribute value combination.

S606: first attribute value combinations, in which the number of commodity categories exceeds a predetermined category threshold and the market capacity does not exceed the predetermined market capacity threshold, are extracted, and the extracted first attribute value combinations are determined to be second attribute value combinations.

S607: and determining corresponding potential values of each second attribute value combination according to the category number of the commodity corresponding to each second attribute value combination and sales information in the preset historical time.

In the above embodiment, step S601 may refer to the description of step S101 shown in fig. 1, the descriptions of steps S201 to S204 shown in fig. 2, the descriptions of steps S301 to S304 shown in fig. 3, and the descriptions of step S401 shown in fig. 4; step S602 may refer to the descriptions of steps S502 to S504 shown in fig. 5, and will not be described herein.

In the above embodiment, for step S603, according to the commodity attribute value combination corresponding to each market segment, the commodity satisfying the attribute value combination condition is extracted, and the effective sales information of the extracted commodity, that is, the commodity sales or sales is counted, and the acquiring manner is specifically shown in step S103 in fig. 1, which is not described herein again.

When calculating the frequency of market segment, the weighting factor of commodity sales information is considered. Specifically, when the attribute values are combined, the occurrence frequency of each attribute value combination is recorded, and then the corresponding sales volume/sales amount of the commodity is taken as an expansion coefficient of the frequency, so that the frequency (index definition freq) of each attribute value combination is calculated in a weighted manner.

The expansion coefficient refers to different influences of different sales/sales in order to cope with the same occurrence frequency.

The expansion coefficient is based on a plurality of frequent attribute value combinations, taking commodity sales as an example, firstly calculating the average sales of the whole frequent attribute value combinations, then using the sales/average sales of a certain attribute combination to = expansion coefficient, and finally obtaining the frequency = occurrence frequency = expansion coefficient.

For step S604, to reduce the amount of subsequent computation, only attribute value combinations with higher frequency, that is, first attribute value combinations, may be extracted, and subsequent filtering of attribute value combinations may be performed based on the first attribute value combinations.

In step S605, the effective commodity number (index definition sku_count) of the commodity corresponding to the combination of the first attribute value is analyzed, and it is noted that the commodity number is SKU (Stock Keeping Unit, minimum stock keeping unit), which is not the commodity stock quantity. And simultaneously, acquiring the market capacity of each first attribute value combination, namely the current effective commodity selling quantity.

For step S606, as a further screening of the attribute value combinations. For the market with the market capacity ordered later, the current commodity quantity of the market is insufficient, but the sales volume/sales amount is higher, which means that the market contribution value (sales volume and sales amount) of single commodity in the market is higher, and the market has higher potential.

Thus, the second attribute value combination of which the market capacity does not exceed the predetermined market capacity threshold and the effective commodity amount is not less than the predetermined commodity amount threshold can be screened out. For example, attribute value combinations are selected in which the market volume is at the last 20% and the number of available goods is not less than 5.

It should be noted that, for the acquisition of the second attribute value combination, there may be other ways, for example, after determining the frequency, the number of the commodity types and the market capacity of each attribute value combination, taking the intersection thereof, selecting attribute value combinations (or sub-markets) with higher market sales/sales but insufficient effective commodity types (or low market capacity), and the result obtained by the intersection is closer to the final "potential market", where the "potential market" has better growth potential and growth space in the future.

For step S607, for calculation of the potential value of each second attribute value combination obtained by screening, refer specifically to the description of step S103 shown in fig. 1, which is not described herein again.

In addition, for the obtained second attribute value combination, a corresponding "market sales-market capacity" coordinate system may be generated, specifically referring to fig. 7, the horizontal axis represents market capacity (number of commodity types), the vertical axis represents market sales, and the bubble size represents market contribution value.

For further optimization, some observation indexes can be added, for example, browsing behaviors of users and conversion rate can be identified and judged.

According to the method provided by the embodiment, the commodity attributes are disassembled and recombined in a machine learning mode, and then the actual commodities are combined for matching, so that a commodity combination set with high frequency is positioned, the rules behind commodities can be effectively observed, and directional production and development suggestions are provided for users.

Referring to fig. 8, a main flow chart of a method for determining a specific commodity potential value according to an embodiment of the present invention is shown, including the following steps:

s801: and acquiring the commodities and corresponding characteristic information, determining the similarity between the commodities according to a preset similarity calculation mode, and determining the commodities with the similarity exceeding a preset similarity threshold as similar commodities.

S802: and verifying the attribute value of each commodity based on a preset verification rule, and extracting the commodity with the attribute value not conforming to the verification rule.

S803: and determining similar commodities associated with the extracted commodities, and acquiring attribute values of the similar commodities.

S804: when the attribute values of the similar commodities accord with the verification rule, replacing the attribute values of the extracted commodities to be the attribute values of the similar commodities.

S805: the attribute to which each attribute value belongs and the category to which each attribute belongs are determined, and the attributes belonging to the same category are divided into a group.

S806: and acquiring commodity operation behavior information of the user under each attribute, combining a preset decision model to acquire the importance of each attribute, and extracting the attribute with the importance exceeding the corresponding preset importance threshold under each category.

S807: and determining the attribute value associated with the extracted attribute of each commodity according to the attribute value of each commodity and the attribute to which each attribute value belongs.

S808: and counting the attribute values of each commodity based on a predetermined statistical model, and determining the occurrence number of each attribute value combination.

S809: and determining the support degree of each attribute value combination according to the occurrence times of each attribute value combination and the number of all commodities, and extracting the attribute value combination with the support degree exceeding a preset support degree threshold.

S810: and acquiring the occurrence times of each attribute value combination, and determining the frequency of each attribute value combination by combining sales information of the commodity corresponding to each attribute value combination in a preset historical time.

S811: and extracting attribute value combinations with the frequency exceeding a preset frequency threshold value, and determining the extracted attribute value combinations as first attribute value combinations.

S812: and obtaining the category number and the market capacity of the commodity corresponding to each first attribute value combination.

S813: first attribute value combinations, in which the number of commodity categories exceeds a predetermined category threshold and the market capacity does not exceed the predetermined market capacity threshold, are extracted, and the extracted first attribute value combinations are determined to be second attribute value combinations.

S814: and determining corresponding potential values of each second attribute value combination according to the category number of the commodity corresponding to each second attribute value combination and sales information in the preset historical time.

The whole process of the invention is as follows: commodity attribute value combination-commodity potential market analysis-commodity attribute value combination in potential market-commodity set matching the attribute value combination, namely the identification from potential market to potential commodity is completed.

As identification of potential goods, further focusing and drill down to "potential markets" can be considered.

In addition, in addition to the technical scheme of the invention, the potential judgment can be carried out on the local market by a method for judging the flow rate acceleration in a short period, for example, whether a commodity is concerned with rising attention in a short period is identified by means of the flow rate, so that the commodity is judged to be potential explosion. However, the judgment method has larger volatility and randomness, and has reference value unlike the scheme of the invention.

The method provided by the embodiment of the invention can be used for supporting upstream users (including suppliers, manufacturers, brands and the like) to provide suggestions and guidance for the on-demand manufacture and development of new products for future products, and for the market segment identified as having sales growth potential, the users can increase the investment of product resources in the market segment. In addition, the fineness of market potential analysis is improved according to the grouping of the attribute values.

Referring to fig. 9, a schematic diagram of main modules of a device 900 for determining a commodity potential value according to an embodiment of the present invention is shown, including:

the attribute value acquisition module 901 is configured to acquire commodities, and determine an attribute value of each commodity;

the attribute value combination module 902 is configured to perform statistics on attribute values of each commodity based on a predetermined statistical model, so as to obtain an attribute value combination;

the potential value determining module 903 is configured to determine, based on the attribute value of each commodity, a commodity corresponding to the attribute value combination, and determine a potential value of the attribute value combination according to the number of types of commodities corresponding to the attribute value combination and sales information in a predetermined history period.

In the embodiment of the present invention, the attribute value obtaining module 901 is further configured to:

Verifying the attribute value of each commodity based on a preset verification rule, and extracting commodities of which the attribute values do not accord with the verification rule; determining similar commodities associated with the extracted commodities, and acquiring attribute values of the similar commodities; and when the attribute values of the similar commodities accord with the verification rule, replacing the attribute values of the extracted commodities to be the attribute values of the similar commodities.

In the embodiment of the present invention, the attribute value obtaining module 901 is further configured to: and acquiring characteristic information of each commodity, obtaining the similarity among the commodities according to a preset similarity calculation mode, and determining the commodities with the similarity exceeding a preset similarity threshold as similar commodities.

In the embodiment of the present invention, the attribute value obtaining module 901 is further configured to: determining the attribute to which each attribute value belongs and the category to which each attribute belongs, and dividing the attributes belonging to the same category into a group;

acquiring commodity operation behavior information of a user under each attribute, combining a preset decision model to acquire the importance of each attribute, and extracting the attribute of which the importance exceeds a corresponding preset importance threshold under each category; and determining the attribute value associated with the extracted attribute of each commodity according to the attribute value of each commodity and the attribute to which each attribute value belongs.

In the embodiment of the present invention, an attribute value obtaining module 901 is configured to: acquiring commodities, determining commodity information of each commodity, and splitting each commodity information based on a preset splitting mode to obtain corresponding attribute values.

In the embodiment of the present invention, the attribute value combination module 902 is configured to: based on a preset statistical model, counting attribute values of each commodity, and determining the occurrence number of each attribute value combination;

and determining the support degree of each attribute value combination according to the occurrence times of each attribute value combination and the number of all commodities, and extracting the attribute value combination with the support degree exceeding a preset support degree threshold.

In the implementation device of the present invention, the potential value determining module 903 is configured to:

acquiring the occurrence times of each attribute value combination, determining the frequency of each attribute value combination by combining sales information of commodities corresponding to each attribute value combination in a preset historical time, extracting attribute value combinations with the frequency exceeding a preset frequency threshold, and determining the extracted attribute value combinations as first attribute value combinations; acquiring the category number and market capacity of the commodities corresponding to each first attribute value combination, extracting first attribute value combinations with the category number exceeding a preset category threshold value and the market capacity not exceeding the preset market capacity threshold value, and determining the extracted first attribute value combinations as second attribute value combinations; and determining corresponding potential values of each second attribute value combination according to the category number of the commodity corresponding to each second attribute value combination and sales information in the preset historical time.

In addition, the specific implementation of the potential value determining apparatus according to the embodiment of the present invention has been described in detail in the above-described method for determining the potential value of the commodity, and thus the description thereof will not be repeated here.

The device provided by the embodiment of the invention can be used for supporting upstream users (including suppliers, manufacturers, brands and the like) to provide suggestions and guidance for the on-demand manufacture and development of new products for future products, and for the market segment identified as having sales growth potential, the users can increase the investment of product resources in the market segment. In addition, the definition of the commodity potential value is improved according to the grouping of the attribute values.

Fig. 10 shows an exemplary system architecture 1000 to which the method of determining a commodity potential value or the apparatus of determining a commodity potential value of the embodiments of the present invention may be applied.

As shown in fig. 10, a system architecture 1000 may include terminal devices 1001, 1002, 1003, a network 1004, and a server 1005 (by way of example only). The network 1004 serves as a medium for providing a communication link between the terminal apparatuses 1001, 1002, 1003 and the server 1005. The network 1004 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

A user can interact with a server 1005 via a network 1004 using terminal apparatuses 1001, 1002, 1003 to receive or transmit messages or the like. Various communication client applications such as shopping class applications, web browser applications, search class applications, instant messaging tools, mailbox clients, social platform software, etc. (by way of example only) may be installed on the terminal devices 1001, 1002, 1003.

The terminal devices 1001, 1002, 1003 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablets, laptop and desktop computers, and the like.

The server 1005 may be a server providing various services, such as a background management server (merely an example) providing support for shopping-type websites browsed by the user using the terminal apparatuses 1001, 1002, 1003. The background management server may analyze and process the received data such as the product information query request, and feedback the processing result (e.g., the target push information, the product information—only an example) to the terminal device.

It should be noted that, the method for determining the commodity potential value provided in the embodiment of the present invention is generally executed by the server 1005, and accordingly, the device for determining the commodity potential value is generally disposed in the server 1005.

It should be understood that the number of terminal devices, networks and servers in fig. 10 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Referring now to FIG. 11, there is illustrated a schematic diagram of a computer system 1100 suitable for use in implementing the terminal device of an embodiment of the present invention. The terminal device shown in fig. 11 is only an example, and should not impose any limitation on the functions and the scope of use of the embodiment of the present invention.

As shown in fig. 11, the computer system 1100 includes a Central Processing Unit (CPU) 1101, which can execute various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 1102 or a program loaded from a storage section 1108 into a Random Access Memory (RAM) 1103. In the RAM 1103, various programs and data required for the operation of the system 1100 are also stored. The CPU 1101, ROM 1102, and RAM 1103 are connected to each other by a bus 1104. An input/output (I/O) interface 1105 is also connected to bus 1104.

The following components are connected to the I/O interface 1105: an input section 1106 including a keyboard, a mouse, and the like; an output portion 1107 including a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, a speaker, and the like; a storage section 1108 including a hard disk or the like; and a communication section 1109 including a network interface card such as a LAN card, a modem, and the like. The communication section 1109 performs communication processing via a network such as the internet. The drive 1110 is also connected to the I/O interface 1105 as needed. Removable media 1111, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like, is installed as needed in drive 1110, so that a computer program read therefrom is installed as needed in storage section 1108.

In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication portion 1109, and/or installed from the removable media 1111. The above-described functions defined in the system of the present invention are performed when the computer program is executed by a Central Processing Unit (CPU) 1101.

The computer readable medium shown in the present invention may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules involved in the embodiments of the present invention may be implemented in software or in hardware. The described modules may also be provided in a processor, for example, as: a processor comprises an attribute value acquisition module, an attribute value combination module and a potential value determination module. The names of these modules do not in some way constitute a limitation of the module itself, for example, the potential value analysis module may also be described as "potential value determination module of attribute value combinations".

As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be present alone without being fitted into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to include:

acquiring commodities and determining attribute values of each commodity;

According to the technical scheme provided by the embodiment of the invention, the method and the device can be used for supporting upstream users (including suppliers, manufacturers, brands and the like) to provide suggestions and guidance for manufacturing and developing new products according to future products, and for the market segment identified as having sales growth potential, the users can increase the investment of product resources in the market segment.

The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives can occur depending upon design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.

Claims

1. A method for determining a commodity potential value, comprising:

acquiring commodities and determining attribute values of each commodity;

based on a preset statistical model, the attribute value of each commodity is counted to obtain an attribute value combination, and the method comprises the following steps: based on a preset statistical model, counting the attribute values of each commodity, and determining the occurrence times of each attribute value combination; determining the support degree of each attribute value combination according to the occurrence times of each attribute value combination and the number of all commodities, and extracting attribute value combinations with the support degree exceeding a preset support degree threshold; the support degree is the quotient of the number of times that different attribute values occur in the attribute value of each commodity and the number of all commodities;

determining the commodity corresponding to the attribute value combination based on the attribute value of each commodity, and determining the potential value of the attribute value combination according to the category number of the commodity corresponding to the attribute value combination and sales information in a preset historical time length, wherein the potential value comprises the following steps:

Acquiring the occurrence times of each attribute value combination, determining the frequency of each attribute value combination by combining sales information of commodities corresponding to each attribute value combination in the preset historical time, extracting attribute value combinations with the frequency exceeding a preset frequency threshold, and determining the extracted attribute value combinations as first attribute value combinations; the frequency is the product of the occurrence frequency and an expansion coefficient, and the expansion coefficient is the quotient of the sales of a certain attribute value combination and the average sales of the whole frequent attribute value combination;

acquiring the category number and market capacity of the commodities corresponding to each first attribute value combination, extracting first attribute value combinations with the category number exceeding a preset category threshold value and the market capacity not exceeding the preset market capacity threshold value, and determining the extracted first attribute value combinations as second attribute value combinations;

and determining corresponding potential values of each second attribute value combination according to the category number of the commodity corresponding to each second attribute value combination and sales information in the preset historical time.

2. The method of claim 1, wherein the acquiring the commodity, determining the attribute value for each commodity, further comprises:

Verifying the attribute value of each commodity based on a preset verification rule, and extracting commodities of which attribute values do not accord with the verification rule;

determining similar commodities associated with the extracted commodities, and acquiring attribute values of the similar commodities;

and when the attribute values of the similar commodities accord with the verification rule, replacing the attribute values of the extracted commodities to be the attribute values of the similar commodities.

3. The method of claim 2, further comprising, prior to said verifying each attribute value based on a predetermined verification rule:

and acquiring characteristic information of each commodity, obtaining the similarity among the commodities according to a preset similarity calculation mode, and determining the commodities with the similarity exceeding a preset similarity threshold as similar commodities.

4. The method of claim 1, wherein the acquiring the commodity, determining the attribute value for each commodity, further comprises:

determining the attribute to which each attribute value belongs and the category to which each attribute belongs, and dividing the attributes belonging to the same category into a group;

acquiring commodity operation behavior information of a user under each attribute, combining a preset decision model to acquire the importance of each attribute, and extracting the attribute of which the importance exceeds a corresponding preset importance threshold under each category;

And determining the attribute value associated with the extracted attribute of each commodity according to the attribute value of each commodity and the attribute to which each attribute value belongs.

5. The method of claim 1, wherein the acquiring the commodity, determining the attribute value for each commodity, comprises:

acquiring commodities, determining commodity information of each commodity, and splitting each commodity information based on a preset splitting mode to obtain corresponding attribute values.

6. A device for determining a commodity potential value, comprising:

the attribute value combination module is configured to perform statistics on attribute values of each commodity based on a predetermined statistical model, to obtain attribute value combinations, and includes: based on a preset statistical model, counting the attribute values of each commodity, and determining the occurrence times of each attribute value combination; determining the support degree of each attribute value combination according to the occurrence times of each attribute value combination and the number of all commodities, and extracting attribute value combinations with the support degree exceeding a preset support degree threshold; the support degree is the quotient of the number of times that different attribute values occur in the attribute value of each commodity and the number of all commodities;

A potential value determining module, configured to determine, based on the attribute value of each commodity, a commodity corresponding to the attribute value combination, and determine, according to the number of types of commodities corresponding to the attribute value combination and sales information in a predetermined history period, a potential value of the attribute value combination, including:

7. The apparatus of claim 6, wherein the attribute value acquisition module is further configured to:

8. The apparatus of claim 7, wherein the attribute value acquisition module is further configured to:

9. The apparatus of claim 6, wherein the attribute value acquisition module is further configured to:

10. The apparatus of claim 6, wherein the attribute value acquisition module is configured to:

11. An electronic device, comprising:

one or more processors;

storage means for storing one or more programs,

when executed by the one or more processors, causes the one or more processors to implement the method of any of claims 1-5.

12. A computer readable medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the method according to any of claims 1-5.