WO2019085372A1 - Commodity recommendation method and commodity recommendation system for online shopping mall - Google Patents

Abstract

Provided are a commodity recommendation method for an online shopping mall, and a commodity recommendation system for an online shopping mall. The commodity recommendation method for an online shopping mall comprises: acquiring data information of any account for logging in to an online shopping mall and shopping mall product information; scoring commodities in the online shopping mall according to the data information and the shopping mall product information, so as to acquire scores of the commodities; and determining a commodity recommendation sequence according to the scores of the commodities. According to the commodity recommendation method provided in the present invention, an analysis is carried out according to multi-dimensional data of users, and multiple dimensions that the users focus on are taken into consideration, such that a recommended commodity is more in line with the requirements of the users, thereby being both practicable and scientific.

Description

Product recommendation method and product recommendation system of online mall

This application claims the priority of the Chinese Patent Application filed on October 31, 2017, the Chinese Patent Office, the application number is 201711048948.4, and the invention name is "the online shopping mall product recommendation method, the product recommendation system", the entire contents of which are incorporated by reference. In this application.

Technical field

The invention relates to the technical field of online shopping malls, in particular to a product recommendation method of an online shopping mall and a product recommendation system of an online shopping mall.

Background technique

The existing mall product recommendation is not fine enough, mainly based on the number of clicks to conduct product recommendation. The user clicks on the number of times a product is viewed. When the user opens the webpage again, the order of display of the product is higher, but the recommendation method is analyzed. There are few dimensions to consider, and it does not really reflect the products that the user really wants. It is not systematic enough. The actual recommended products may not be what the user really wants.

Therefore, how to realize the order of goods recommended by the online mall is more in line with the needs of users has become an urgent problem to be solved.

Summary of the invention

The present invention aims to solve at least one of the technical problems existing in the prior art or related art.

To this end, the first aspect of the present invention provides a method for recommending products for an online mall.

A second aspect of the present invention is to provide a product recommendation system for an online shopping mall.

In view of this, according to an aspect of the present invention, a method for recommending an item of an online shopping mall is provided, comprising: obtaining data information of any account registered in the online shopping mall and product information of the shopping mall; and online shopping according to the data information and the product information of the shopping mall. The product is scored to obtain the product score; the product recommendation order is determined based on the product score.

The product recommendation method of the online shopping mall provided by the present invention, when a user logs in to the online mall, obtains the information of the user account, and then obtains the data information of the account, the data information reflects the operation information, feedback information, etc. of the user in the online mall, and obtains at the same time Product information in the mall, such as the total amount of a certain product, according to the obtained data information and the product information of the mall, the products in the online mall are scored, and the scores of all the products are obtained, because the scoring process is based on different operations of each user in the mall. Information and feedback information are used for data, so different users have different scores for the products in the mall, and the product recommendation order is determined according to the product score. The higher the score, the higher the recommendation order, and the individuality for different users. The customization method of the present invention is not limited to the number of clicks, and includes various consideration dimensions such as user operation information and feedback information. In addition, the recommendation method of the present invention is related to the product product data in the mall, and can be implemented. Different total number of items recommended in different order In order to make the recommended products more in line with the needs of users, on the other hand, it is more conducive to the operation and maintenance operations of the mall, more practical and scientific.

According to the product recommendation method of the above online shopping mall according to the present invention, the following technical features may also be provided:

In the above technical solution, preferably, before acquiring the data information of the online shopping mall, the method further includes: determining a preset rating dimension of the category product according to the type of the commodity; and storing a correspondence between the preset rating dimension and the category commodity.

In the technical solution, before each user logs in to the online shopping mall, the preset rating dimension of the product is determined in advance according to the type of the product, and the preset rating dimensions of different types of products are different, and the corresponding relationship between the rating dimension of each product and the product is stored. In order to allow users to perform preset dimension operations or feedback information for all products, and then perform product ratings.

In any one of the foregoing technical solutions, preferably, the product in the online shopping mall is scored according to the data information and the shopping mall product information, so as to obtain the product score, which includes: obtaining any account to pre-order any product on the online shopping mall. The evaluation information of the rating dimension is set; according to the preset rule, the evaluation information of the preset rating dimension is converted into the score of the corresponding dimension; the product information of the mall of any commodity is obtained; and the mall product information of any commodity is converted according to the preset rule. The score for the corresponding dimension; the product of each corresponding dimension is multiplied by the weight of the corresponding dimension to obtain the product score for any item.

In the technical solution, after the user logs in to the online shopping mall, the operation of the goods in the shopping mall can be performed, the operation times can be converted into evaluation information, the feedback dimension is fed back to the evaluation information, and the user account is obtained for the product evaluation information on the online shopping mall, according to The rules preset by the mall convert these evaluation information into the scores of the dimension, and then multiply the scores of each dimension by the weights of the dimensions, and the weights of the important or more important dimensions of the user are set higher, not so important. Or the dimension weight setting of less attention is lower, and then the mall product information of any commodity is obtained, and then the score of the corresponding dimension is obtained according to the mall product information of any commodity, and finally the total score of the commodity is obtained as the commodity score of the commodity. . In this way, different evaluation information of different users affects the product score, and different user attention dimensions are different for setting the dimension weights, so ultimately each user sees different recommendation orders, more suitable for each user's needs, and achieves personalized customization; The final score of the product is also related to the product information in the mall, and the products that are more favorable to the operation and maintenance of the mall are preferred.

In any one of the foregoing technical solutions, before the obtaining the evaluation information of the preset dimension of any item on the online shopping mall, the method further includes: determining whether the evaluation information of any account is received; when receiving the evaluation information The step of obtaining the evaluation information of the preset dimension of any item on the online mall by any account; when not receiving the evaluation information of any account, obtaining the account according to any account other than any account Evaluate all product scores from the information, and use the average of all product scores as the product score.

In the technical solution, before obtaining the evaluation information of the preset dimension of any item on the online mall by any account, first determining whether the user account has received the evaluation information of a certain product, and if the user has conducted the evaluation, then selecting the The user account evaluates the product, and the product score is confirmed based on the personal evaluation information of the user account. If the user account has not received the evaluation information of the product, the user account number other than the user account is obtained. The product score derived from the evaluation information of this product, and the average of the product scores of all user accounts is taken as the product score of the product corresponding to the user account.

In any of the above technical solutions, preferably, the preset scoring dimension includes at least: a commodity price, a number of returns, a click count, a purchase count, a product quality, a product attribute, and a total quantity of the goods.

In this technical solution, the limited preset rating dimension is not only the number of clicks, but also the commodity price, the number of returns, the number of purchases, the quality of the goods, and the attributes of the goods. The more clicks and purchases, the higher the score; the higher the price of the product, the lower the score; the higher the quality of the product, the higher the score, the more the product attributes meet the user's needs, the higher the score; the higher the total quantity, the higher the score .

According to a second aspect of the present invention, a product recommendation system for an online shopping mall is provided, comprising: a first obtaining unit, configured to obtain data information of any account of the online shopping mall and product information of the shopping mall; The data information and the product information of the mall are used to score the products in the online shopping mall to obtain the product score; the recommendation unit is used to determine the order of product recommendation according to the product score.

The product recommendation system of the online shopping mall provided by the present invention, when a certain user logs in to the online shopping mall, the first obtaining unit obtains the information of the user account, and then obtains the data information of the account, and the data information reflects the operation information of the user in the online mall, and the feedback information Waiting for the product information in the mall, such as the total amount of a certain product, the scoring unit scores the products in the online mall according to the obtained data information and the product information of the mall, and obtains the scores of all the products, because the scoring process is based on each user. The data of different operation information and feedback information in the mall is performed, so different users have different scores for the products in the mall, and the recommendation unit determines the order of product recommendation according to the product scores, and the higher the score, the more recommended the order Before the implementation of personalized customization for different users, the recommendation method of the present invention is not limited to the number of clicks, and includes various consideration dimensions such as user operation information and feedback information, and the recommendation method of the present invention is in the mall. Commodity product data related, can be achieved Unlike the total amount of goods recommended order so that recommended product more in line with user needs on the one hand, on the other hand is more conducive to the operation and maintenance operation Mall, more practical and scientific.

According to the product recommendation system of the above online shopping mall according to the present invention, the following technical features may also be provided:

In the above technical solution, preferably, the determining unit is configured to determine a preset rating dimension of the category item according to the type of the commodity; and the storage unit is configured to store the correspondence between the preset rating dimension and the category item.

In the technical solution, before each user logs in to the online shopping mall, the determining unit determines the preset rating dimension of the commodity according to the type of the commodity in advance, and the preset rating dimension of the different types of commodities is different, and the storage unit stores the rating dimension of each commodity and The corresponding relationship of the products enables the user to perform operation or feedback information of preset dimensions for all the products, and then perform product evaluation.

In any one of the above technical solutions, preferably, the scoring unit comprises: a second obtaining unit, configured to obtain evaluation information of a preset scoring dimension of any item on the online mall; According to a preset rule, the evaluation information of the preset rating dimension is converted into the score of the corresponding dimension; the third obtaining unit is configured to obtain the mall product information of any commodity; and the second converting unit is configured to The product information of the commodity is converted into the score of the corresponding dimension; the calculation unit is configured to multiply the score of each corresponding dimension by the weight of the corresponding dimension to obtain the commodity score of any commodity.

In the technical solution, after the user logs in to the online mall, the goods in the mall are operated, the number of operations can be converted into evaluation information, the feedback dimension is fed back to the evaluation information, and the second obtaining unit obtains the user account to the online mall. The evaluation information, the first conversion unit converts the evaluation information into the score of the dimension according to the preset rules of the mall, and then the calculation unit multiplies the score of each dimension by the weight of the dimension, which is important to the user or more concerned. The weight of the dimension is set higher, the dimension weight of the less important or less concerned is set lower, and then the third obtaining unit acquires the mall product information of any commodity, and the second converting unit obtains the product information of the mall according to any commodity. The score of the corresponding dimension, and finally the total score of the product, as the product score of the product. In this way, different evaluation information of different users affects the product score, and different user attention dimensions are different for setting the dimension weights, so ultimately each user sees different recommendation orders, more suitable for each user's needs, and achieves personalized customization; The final score of the product is also related to the product information in the mall, and the products that are more favorable to the operation and maintenance of the mall are preferred.

In any one of the foregoing technical solutions, the determining unit is configured to determine whether the evaluation information of any account is received, and the second obtaining unit is configured to acquire any account for the online mall when receiving the evaluation information. The step of evaluating information of the preset dimension of any of the products; the averaging unit is configured to obtain, according to the evaluation information of any account other than any account, the evaluation information of any commodity For all product scores, the average of all product scores is taken as the product score.

In the technical solution, before obtaining the evaluation information of the preset dimension of any item on the online mall by any account, the determining unit first determines whether the user account has received the evaluation information of a certain product, and if the user has conducted the evaluation, then The second obtaining unit selects the evaluation information of the user account for the product, and confirms the product score based on the personal evaluation information of the user account. If the user account has not received the evaluation information of the product, the average unit acquires the user account. The product scores obtained by the other evaluated user accounts for the evaluation information of the product, and the average of the product scores of all the user accounts is taken as the product score of the product corresponding to the user account.

In any of the above technical solutions, preferably, the preset scoring dimension includes at least: a commodity price, a number of returns, a click count, a purchase quantity, a product quality, a commodity attribute, and a total quantity of the goods.

In this embodiment, the defined preset rating dimension is not only the number of clicks, but also the price of the item, the number of returns, the number of purchases, the quality of the item, and the attribute of the item. The more clicks and purchases, the higher the score; the higher the price of the product, the lower the score; the higher the quality of the product, the higher the score, the more the product attributes meet the user's needs, the higher the score; the higher the total quantity, the higher the score .

Additional aspects and advantages of the invention will be apparent from the description of the invention.

DRAWINGS

The above and/or additional aspects and advantages of the present invention will become apparent and readily understood from

1 is a schematic flow chart showing a product recommendation method of an online shopping mall according to an embodiment of the present invention;

2 is a schematic flow chart showing a product recommendation method of an online shopping mall according to another embodiment of the present invention;

3 is a schematic block diagram of a product recommendation system of an online shopping mall according to an embodiment of the present invention;

4 is a schematic block diagram of a merchandise recommendation system of an online shopping mall in accordance with another embodiment of the present invention.

Detailed ways

The present invention will be further described in detail below with reference to the drawings and specific embodiments. It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict.

In the following description, numerous specific details are set forth in order to facilitate a full understanding of the invention, but the invention may be practiced in other embodiments than described herein. Limitations of the embodiments.

An embodiment of the first aspect of the present invention provides a product recommendation method for an online shopping mall. FIG. 1 is a schematic flow chart of a product recommendation method for an online shopping mall according to an embodiment of the present invention:

Step 102: Obtain data information of any account registered in the online mall and product information of the mall;

Step 104: Rate the products in the online mall according to the data information and the product information of the mall to obtain the product score;

Step 106: Determine a product recommendation order according to the product score.

The product recommendation method of the online mall in this embodiment, when a user logs in to the online mall, obtains the information of the user account, and then obtains the data information of the account, the data information reflects the operation information, feedback information, etc. of the user in the online mall, and simultaneously acquires Product information in the mall, such as the total amount of a certain product, according to the obtained data information and the product information of the mall, the products in the online mall are scored, and the scores of all the products are obtained, because the scoring process is based on different operations of each user in the mall. Information and feedback information are used for data, so different users have different scores for the products in the mall, and the product recommendation order is determined according to the product score. The higher the score, the higher the recommendation order, and the individuality for different users. The customization method of the present invention is not limited to the number of clicks, and includes various consideration dimensions such as user operation information and feedback information. In addition, the recommendation method of the present invention is related to the product product data in the mall, and can be implemented. Different total number of items are recommended in different order. Recommendable product more in line with user needs on the one hand, on the other hand is more conducive to the operation and maintenance operation Mall, more practical and scientific.

FIG. 2 is a flow chart showing a method for recommending a product in an online shopping mall according to another embodiment of the present invention. Among them, the product recommendation methods include:

Step 202: Determine a preset rating dimension of the category product according to the type of the commodity;

Step 204: Store a correspondence between a preset rating dimension and a category product;

Step 206: Obtain data information of any account that is logged in to the online mall;

Step 208, it is determined whether the evaluation information of any account is received, if not, then proceeds to step 210, and then proceeds to step 212;

Step 210: Acquire all the product scores obtained from the evaluation information of any commodity other than any account, and use the average value of all the commodity scores as the commodity score;

Step 212: Obtain evaluation information of a preset rating dimension of any item on the online mall by any account;

Step 214: Convert the evaluation information of the preset scoring dimension into a score of the corresponding dimension according to a preset rule;

Step 216, obtaining mall product information of any commodity;

Step 218: Convert the mall product information of any commodity into a score of the corresponding dimension according to a preset rule;

Step 220, multiplying the score of each corresponding dimension by the weight of the corresponding dimension to obtain the commodity score of any commodity;

Step 222, determining a product recommendation order according to the product score.

In this embodiment, before each user logs in to the online shopping mall, the preset rating dimension of the product is determined in advance according to the type of the product, and the preset rating dimension includes at least: the product price, the number of returns, the number of clicks, the number of purchases, the quality of the product, Product attributes. The limited default rating dimension is more than the number of clicks, but also includes the price of the item, the number of returns, the number of purchases, the quality of the item, the attribute of the item, and the total quantity of the item. The more clicks and purchases, the higher the score; the higher the price of the product, the lower the score; the higher the quality of the product, the higher the score, the more the product attributes meet the user's needs, the higher the score; the higher the total quantity, the higher the score . Different types of products have different preset scoring dimensions. For example, the preset scoring dimension of vegetables is freshness, price, aesthetics, nutritional value, yield (the total amount of certain vegetables in the mall / the number of active users in the mall), the number of clicks, and the purchase. The number of times, the number of returns, the vegetables are scored according to these 8 dimensions. For the dimension of a certain vegetable yield, the value obtained by dividing the total amount of the vegetable in the mall by the number of currently active users in the mall is used as a criterion for evaluating the number of yields. The higher the value, the higher the score of the dimension, for example, The current active population in the mall is 500 people, the total amount of tomatoes is 1000 kg, the total amount of broccoli is 2000 kg, then the output of tomatoes is 1000/500, the average person is 2 kg, and the broccoli is 4 kg per person, then When the mall scores consistently for other dimensions, it is preferred to recommend broccoli with a high yield, and vegetables with a large total amount of vegetables are preferentially sold. Finally, the correspondence between the rating dimension of each item and the item is stored, so that the user performs the operation or feedback information of the preset dimension for all the items, and then performs the product rating.

In this embodiment, after the user logs in to the online shopping mall, the operation of the goods in the shopping mall can be performed, the operation times can be converted into evaluation information, and the feedback dimension is fed back to the evaluation information, and the user account is obtained for the product evaluation information on the online shopping mall, according to The rules preset by the mall convert these evaluation information into scores of this dimension, such as tomatoes: the new dimension can be divided into: A is very fresh, score 1, B is relatively fresh, score is 0.8, C gloss is dim, not very fresh, The score is 0.4, D is very poor, there are signs of decay, the score is 0.2, giving the weight of each dimension, and then the score of each dimension is multiplied by the weight of the dimension, and the weight of the important or more important dimension of the user is set. The higher, less important or less concerned dimension weight setting is lower, then the third obtaining unit obtains the mall product information of any commodity, and the second converting unit obtains the score of the corresponding dimension according to the mall product information of any commodity. , and finally get the total score of the product as the product score of the product. For example: fresh dimension score 0.6, taste dimension score 0.5, price dimension score 0.4, aesthetic dimension score 0.5, nutritional value dimension score 0.5, yield score 0.5, click score 0.6, purchase count 0.4, return count 0.5, weight respectively Go to 1,1,1,0.8,1,1,1,1, so the tomato scores 0.6×1+0.5×1+0.4×1+0.5×0.8+0.5×1+0.5×1+0.6×1+0.4× 1+0.5×1=4.4. In this way, different evaluation information of different users affects the product score, and different user attention dimensions are different for setting the dimension weights, so ultimately each user sees different recommendation orders, more suitable for each user's needs, and achieves personalized customization; The final score of the product is also related to the product information in the mall, and the products that are more favorable to the operation and maintenance of the mall are preferred.

In this embodiment, before obtaining the evaluation information of any account on the preset dimension of any item on the online mall, first determining whether the user account has received evaluation information on an item, and if the user has conducted the evaluation, then selecting the The user account evaluates the product, and the product score is confirmed based on the personal evaluation information of the user account. If the user account has not received the evaluation information of the product, the user account number other than the user account is obtained. The product score derived from the evaluation information of this product, and the average of the product scores of all user accounts is taken as the product score of the product corresponding to the user account. That is to say, if there is a personal evaluation of the individual, if the individual has no value, take the average, for example: Zhang bought a packet of tomatoes, and gave feedback on the tomato, in the fresh, taste, price, beauty, nutrition Value and other dimensions have been evaluated, then Zhang’s tomato scores take his own evaluation score. If Zhang did not buy broccoli, then Zhang’s broccoli score is the average score of all users on broccoli, and so on, there will be a score for all dishes, and then sorted from big to small, this is According to this sorting rule, the recommended order that each user sees should be different.

An embodiment of the second aspect of the present invention provides a product recommendation system 300 for an online shopping mall. FIG. 3 is a schematic block diagram of a product recommendation system 300 for an online shopping mall according to an embodiment of the present invention. As shown in FIG. 3, the product recommendation system 300 of the online mall includes an acquisition unit 10, a scoring unit 12, and a recommendation unit 14. The obtaining unit 10 is configured to obtain data information of any account registered in the online mall and product information of the mall; the scoring unit 12 is configured to score the products in the online mall according to the data information and the product information of the mall to obtain the product score; The unit 14 is configured to determine a product recommendation order according to the product score.

In the product recommendation system 300 of the online shopping mall in this embodiment, when a certain user logs in to the online shopping mall, the obtaining unit 10 obtains the information of the user account, and then obtains the data information of the account, and the data information reflects the operation information of the user in the online mall, and the feedback Information, etc., while obtaining product information in the mall, such as a certain product total, the scoring unit 12 scores the products in the online mall according to the obtained data information and the mall product information, and obtains the scores of all the products, because the scoring process is based on each The users perform different data such as operation information and feedback information in the mall, so different users have different scores for the products in the mall, and the recommendation unit 14 determines the order of product recommendation according to the product scores, and the products with higher scores are recommended. The higher the order, the more personalized customization for different users, and the recommendation method of the present invention is not limited to the number of clicks, and includes various operational dimensions such as user operation information and feedback information. In addition, the recommendation method of the present invention Relevant to the product data in the mall, the needle can be realized The total number of different different product recommendation order so that recommended product more in line with user needs on the one hand, on the other hand is more conducive to the operation and maintenance operation Mall, more practical and scientific.

4 is a schematic block diagram of a merchandise recommendation system 400 of an online shopping mall in accordance with another embodiment of the present invention. As shown in FIG. 4, the product recommendation system 400 of the online mall includes a first acquisition unit 20, a scoring unit 22, a recommendation unit 24, a determination unit 26, a storage unit 28, a determination unit 30, and an averaging unit 32. The scoring unit 22 specifically includes: a second obtaining unit 220, a first converting unit 222, a third obtaining unit 224, a second converting unit 226, and a calculating unit 228.

The first obtaining unit 20 is configured to obtain data information of any account of the online shopping mall; the scoring unit 22 is configured to score the products in the online shopping mall according to the data information to obtain the product score; and the recommending unit 24 is configured to The item score determines the item recommendation order; the determining unit 26 is configured to determine a preset rating dimension of the category item according to the type of the item; the storage unit 28 is configured to store a correspondence between the preset rating dimension and the category item; the determining unit 30 is configured to: Determining whether the evaluation information of any account is received; the averaging unit 32 is configured to obtain, according to the evaluation information of any account other than any account, the evaluation information of any account All the product scores, the average of all the product scores is taken as the product score; the second obtaining unit 220 is specifically configured to, when receiving the evaluation information, perform the evaluation information of obtaining the preset dimension of any of the products on the online mall by any account. a conversion unit 222, configured to convert the evaluation information of the preset rating dimension into a corresponding dimension according to a preset rule The third obtaining unit 224 is configured to obtain the mall product information of any commodity; the second converting unit 226 is configured to convert the mall product information of any commodity into the score of the corresponding dimension according to the preset rule; 228. Use to multiply the score of each corresponding dimension by the weight of the corresponding dimension to obtain the product score of any commodity.

In the product recommendation system 400 of the online shopping mall in this embodiment, when a certain user logs in to the online shopping mall, the first obtaining unit 20 acquires the information of the user account, and then obtains the data information of the account, and the data information reflects the operation information of the user in the online mall. The feedback unit 22 scores the products in the online shopping mall according to the obtained data information, and obtains the scores of all the products, because the rating process is based on each user's different operation information and feedback information in the shopping mall. Therefore, different users have different scores for the products in the mall, and the recommendation unit 24 determines the order of product recommendation according to the product scores. The higher the score, the higher the recommendation order, and the personalized customization for different users. The recommended method of the invention considers that the dimension of the product is not limited to the number of clicks, and includes various consideration dimensions such as the user's operation information and feedback information, and the recommended products are more in line with the user's needs, and are more practical and scientific.

In this embodiment, before each user logs in to the online shopping mall, the determining unit 26 determines the preset rating dimension of the commodity according to the type of the commodity in advance, and the preset rating dimension includes at least: the commodity price, the number of returns, the number of clicks, the number of purchases, Product quality, product attributes. The limited default rating dimension is more than the number of clicks, but also includes the price of the item, the number of returns, the number of purchases, the quality of the item, the attribute of the item, and the total quantity of the item. The more clicks and purchases, the higher the score; the higher the price of the product, the lower the score; the higher the quality of the product, the higher the score, the more the product attributes meet the user's needs, the higher the score; the higher the total quantity, the higher the score . Different types of products have different preset scoring dimensions. For example, the default scoring dimension of vegetables is freshness, price, aesthetics, nutritional value, yield (a certain amount of vegetables in the mall / number of active users in the mall), clicks, purchases The number of times, the number of returns, the vegetables are scored according to these 8 dimensions. For the dimension of a certain vegetable yield, the value obtained by dividing the total amount of the vegetable in the mall by the number of currently active users in the mall is used as a criterion for evaluating the number of yields. The higher the value, the higher the score of the dimension, for example, The current active population in the mall is 500 people, the total amount of tomatoes is 1000 kg, the total amount of broccoli is 2000 kg, then the output of tomatoes is 1000/500, the average person is 2 kg, and the broccoli is 4 kg per person, then When the mall scores consistently for other dimensions, it is preferred to recommend broccoli with a high yield, and vegetables with a large total amount of vegetables are preferentially sold. Finally, the storage unit 28 stores the correspondence relationship between the rating dimension of each item and the item, so that the user performs the operation or feedback information of the preset dimension for all the items, thereby performing the product rating.

In this embodiment, after the user logs in to the online shopping mall, the operation of the goods in the shopping mall is performed, the operation times can be converted into evaluation information, and the feedback dimension is fed back to the evaluation information, and the second obtaining unit 220 obtains the user account on the online shopping mall. The commodity evaluation information, the first conversion unit 222 converts the evaluation information into the score of the dimension according to the preset rules of the mall, such as the tomato: the fresh dimension, which can be divided into: A is very fresh, the score is 1, the B is relatively fresh, and the score is 0.8, C gloss is dim, not very fresh, the score is 0.4, D is very poor, there are signs of decay, the score is 0.2, giving the weight of each dimension, and then the calculation unit 228 multiplies the score of each dimension by the weight of the dimension. The weight setting of the dimension that is important to the user or more concerned is higher, the dimension weight setting that is less important or less concerned is lower, and then the third obtaining unit 224 acquires the mall product information of any commodity, and the second conversion unit 226 then obtain the score of the corresponding dimension according to the product information of the mall of any commodity, and finally obtain the total score of the commodity, Commodity fraction of the merchandise. For example: fresh dimension score 0.6, taste dimension score 0.5, price dimension score 0.4, aesthetic dimension score 0.5, nutritional value dimension score 0.5, yield score 0.5, click score 0.6, purchase count 0.4, return count 0.5, weight respectively Go to 1,1,1,0.8,1,1,1,1, so the tomato scores 0.6×1+0.5×1+0.4×1+0.5×0.8+0.5×1+0.5×1+0.6×1+0.4× 1+0.5×1=4.4. In this way, different evaluation information of different users affects the product score, and the different user attention dimensions are different in setting the weight of the dimension. Therefore, each user sees a different recommendation order, which is more suitable for each user's needs, and achieves personalized customization.

In this embodiment, before obtaining the evaluation information of the preset dimension of any item on the online shopping mall, the determining unit 30 determines whether the user account has received the evaluation information of a certain product, and if the user has conducted the evaluation, Then, the second obtaining unit 220 selects the evaluation information of the product by the user account, and confirms the product score based on the personal evaluation information of the user account. If the user account has not received the evaluation information of the product, the averaging unit 32 acquires the user. The product score derived from the evaluation information of the user account other than the account, and the average product score of all the user accounts is taken as the product score of the product corresponding to the user account. That is to say, if there is a personal evaluation of the individual, if the individual has no value, take the average, for example: Zhang bought a packet of tomatoes, and gave feedback on the tomato, in the fresh, taste, price, beauty, nutrition Value and other dimensions have been evaluated, then Zhang’s tomato scores take his own evaluation score. If Zhang did not buy broccoli, then Zhang’s broccoli score is the average score of all users on broccoli, and so on, there will be a score for all dishes, and then sorted from big to small, this is According to this sorting rule, the recommended order that each user sees should be different.

In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "specific embodiments" and the like means that the specific features, structures, materials, or characteristics described in connection with the embodiments or examples are included in the present invention. At least one embodiment or example. In the present specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples.

The above description is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and scope of the present invention are intended to be included within the scope of the present invention.

Claims (10)

A product recommendation method for an online mall, characterized in that it comprises: Obtaining data information and mall product information of any account registered in the online mall; And ranking the goods in the online shopping mall according to the data information and the shopping mall product information to obtain a product score; A recommendation order of the items is determined based on the item score. The method for recommending the product of the online shopping mall according to claim 1, wherein before the obtaining the data information of the online shopping mall and the product information of the shopping mall, the method further comprises: Determining a preset scoring dimension of the category of merchandise according to the type of the merchandise; And storing a correspondence between the preset rating dimension and the category item. The method for recommending the product of the online shopping mall according to claim 2, wherein the product in the online shopping mall is scored according to the data information and the product information of the mall to obtain a product score, which specifically includes: Obtaining evaluation information of the preset rating dimension of any of the items on the online mall by the any account; And converting the evaluation information of the preset rating dimension into a score of the corresponding dimension according to a preset rule; acquiring the shopping mall product information of the any commodity; Converting the mall product information of the any of the commodities into a score of a corresponding dimension according to the preset rule; The score of each of the respective dimensions is multiplied by the weight of the corresponding dimension to obtain the commodity score of the any of the commodities. The method for recommending the product of the online shopping mall according to claim 3, wherein before the obtaining the evaluation information of the preset dimension of any of the products on the online shopping mall, the method further includes: Determining whether the evaluation information of any of the accounts is received; When receiving the evaluation information, performing the step of acquiring the evaluation information of the preset dimension of any of the items on the online mall by the any account; When the evaluation information of the any account is not received, all the product scores obtained according to the evaluation information of the any account other than the account number are obtained, and all the goods are all The average of the product scores is taken as the product score of any of the products. The product recommendation method of the online shopping mall according to any one of claims 2 to 4, wherein the preset rating dimension comprises at least: a product price, a number of returns, a number of clicks, a number of purchases, a quality of the product, and an attribute of the product. The total amount of goods. An item recommendation system for an online mall, characterized in that it comprises: a first obtaining unit, configured to acquire data information and a mall product information of any account of the online shopping mall; a scoring unit, configured to score the products in the online mall according to the data information and the mall product information, to obtain a product score; a recommending unit for determining a recommended order of the items based on the item score. The product recommendation system of the online shopping mall according to claim 6, further comprising: a determining unit, configured to determine a preset scoring dimension of the category of merchandise according to the type of the merchandise; a storage unit, configured to store a correspondence between the preset rating dimension and the category item. The merchandise recommendation system of the online shopping mall according to claim 7, wherein the scoring unit specifically comprises: a second obtaining unit, configured to obtain, by the any account, evaluation information of a preset rating dimension of any item on the online shopping mall; a first conversion unit, configured to convert the evaluation information of the preset rating dimension into a score of a corresponding dimension according to a preset rule; a third obtaining unit, configured to acquire the mall product information of any of the commodities; a second converting unit, configured to convert the mall product information of the any commodity into a score of a corresponding dimension according to the preset rule; a calculating unit, configured to multiply a score of each of the corresponding dimensions by a weight of the corresponding dimension to obtain the commodity score of the any commodity. The product recommendation system of the online shopping mall according to claim 8, further comprising: a determining unit, configured to determine whether the evaluation information of the any account is received; The second obtaining unit is specifically configured to: when receiving the evaluation information, perform the step of acquiring, by the any account, evaluation information of a preset dimension of any commodity on the online mall; An averaging unit, configured to acquire, when the rating information of the any account is not received, all the product scores obtained by evaluating the evaluation information of any of the commodities according to the account other than the account number The average value of all the commodity scores is used as the commodity score of any of the commodities. The product recommendation system of the online shopping mall according to any one of claims 7 to 9, wherein the preset rating dimension comprises at least: a product price, a return quantity, a click quantity, a purchase quantity, a product quality, and a product attribute. The total amount of goods.

Images