Automatic method for virtual goods to go on and off the shelves based on an online live streaming platform

Abstract

The present invention discloses a method for automatically putting virtual goods on and off shelves based on a webcast platform, which relates to the technical field of information processing. The present invention obtains the sales records of the goods sold by the merchant on the webcast platform, marks the remaining goods as historical goods after eliminating redundant goods; obtains the information of the goods sold by the merchant in this session and marks the corresponding goods as on-sale goods; obtains the sell-out ratio and selling profit of the goods according to the information of the goods sold in this session, and calculates the selling value according to the sell-out ratio and selling profit; obtains the attention degree and re-shelving value of the goods according to the extraction of webcast dialogue keywords; combines the attention degree, re-shelving value and selling value to put historical goods on shelves or take on-sale goods off shelves, achieving the purpose of being able to timely analyze various interaction information of users on the webcast platform, reminding the merchant to adjust the status of the goods and ensuring the profit obtained by the maker.

Description

Technical Field

[0001] The present invention belongs to the technical field of information processing, and particularly relates to a method for automatically putting virtual goods on and off shelves based on an online live streaming platform. Background Art

[0002] With the rapid development of e-commerce platforms, the purchasing behaviors of many users and the sales behaviors of merchants have shifted online. On some online shopping platforms, goods are also sold in combination with live streaming.

[0003] Live streaming with goods has become the main sales method on current e-commerce platforms. In live streaming with goods, the host needs to explain the goods, and the staff needs to perform operations such as putting the goods on the shelves and pushing them. For example, Chinese Patent CN113992993A provides a method and system for processing live streaming commodity prompts and shelf information. After the plugin terminal monitors the first message topic, it parses to obtain the store ID and commodity ID, and retrieves the corresponding commodity in the corresponding store as the commodity to be put on the shelf; after detecting the voice shelf instruction, it generates a second message topic for broadcasting; after the plugin terminal monitors the second message topic, it performs the merchant operation of the commodity to be put on the shelf, which is used to manually process the shelf instruction operation of the host.

[0004] Another example is that Chinese Patent CN109978665A provides a method, device, terminal device and storage medium for putting goods on and off shelves. According to the commodity information of the goods put on the shelves by the maker, the price of the goods is identified; the price is compared with a preset commodity price threshold to determine the status information of the goods, where the status information includes the shelf status information or the off-shelf status information; when the status information is the shelf status information, control the goods to be put on the shelf, where putting on the shelf includes displaying the commodity information to the user through a page; when the status information is the off-shelf status information, control the goods to be taken off the shelf, and prompt the maker with the information that the goods have been taken off the shelf, where taking off the shelf includes removing the goods from the currently displayed page. The purpose of being able to remind the maker to adjust the information of the goods and ensure the profit obtained by the maker is achieved. Also, Chinese Patents CN111161028A, CN102024224A, etc., all provide a technical solution for putting goods on and off shelves. However, based only on factors such as the price, trading volume, and number of visitors of the goods, the selection of putting on and off shelves cannot be made in real time according to the demand of the goods. Therefore, a method for automatically putting virtual goods on and off shelves based on an online live streaming platform is provided. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for automatically putting virtual goods on and off shelves based on an online live streaming platform. By combining the analysis of the sales records of the goods, obtaining the attention, restocking value, and selling value of the goods, putting historical goods on the shelves or taking off the goods on sale, the problem of the single method for putting on and off shelves in the existing data processing is solved.

[0006] To solve the above technical problems, the present invention is realized through the following technical solutions:

[0007] The present invention is a method for automatically putting virtual goods on and off the shelves based on an online live streaming platform, including the following steps:

[0008] Step 1: Obtain the sales records of the goods sold by the merchant on the live streaming platform, and mark the remaining goods as historical goods after eliminating redundant goods;

[0009] Step 2: Obtain the information of the goods sold by the merchant in this session, and mark the corresponding goods as on-sale goods;

[0010] Step 3: Obtain the sell-out ratio and selling profit of the goods according to the information of the goods sold in this session, and calculate the selling value according to the sell-out ratio and selling profit;

[0011] Step 4: Conduct a past search: Obtain the attention and restocking value of the goods according to the extraction of live streaming dialogue keywords;

[0012] Step 5: Combine the attention, restocking value, and selling value to put historical goods on the shelves or take off-sale goods off the shelves.

[0013] Furthermore, the goods sales records include the goods name and the unit price, negative review rate, sales volume, listing time, and selling time corresponding to each live stream. The method for obtaining the historical goods is as follows:

[0014] Obtain the listing time, selling time, negative review rate, and profit of the goods from the goods sales records;

[0015] Calculate the selling duration Stj, where Ssij is the listing time of product j in the i-th live stream, Scij is the selling time of product j in the i-th live stream, i = 1, 2, 3,..., n; n ≥ 15; Scjmax and Ssjmin are the longest and shortest times elapsed from listing to selling of product j in n live streams; j = 1, 2, 3,..., m; m is a positive integer;

[0016] Obtain the recommended resistance value Tzj of product j, Cj is the negative review rate of product j after being sold in n live streams, and Lj is the average profit of product j in n live streams;

[0017] Mark the products whose recommended resistance values appear in the last five for three consecutive times as redundant products, and the rest as historical products.

[0018] Furthermore, the method for obtaining the sell-out ratio of the goods according to the information of the goods sold in this session is as follows:

[0019] Optionally select one on-sale product and mark it as the target product;

[0020] Obtain the total number of items, the listing time, and the sold-out time of the target product;

[0021] With the help of the sold-out time and the listing time, obtain the sales duration of the target product, where the sales duration = sold-out time - listing time;

[0022] With the help of the sales duration and the total number of items of the target product, obtain the sell-through ratio. Among them, the discount of the target product in this live broadcast is β;

[0023] If the target product is not sold out in this live broadcast, the sell-through ratio = listing duration / number of items sold;

[0024] The sales duration and the listing duration are calculated in minutes.

[0025] Furthermore, the method for obtaining the selling profit of the product according to the product information sold in this session is as follows:

[0026] S1: Calculate the gross profit of the target product, where the gross profit = selling unit price - cost;

[0027] S2: The selling profit = number of items sold * gross profit - sell-through ratio * number of items sold * α, where α is the live broadcast cost per minute.

[0028] Furthermore, calculate the selling value according to the sell-through ratio and the selling profit, where the selling value = 0.532 * sell-through ratio + 0.468 * selling profit, and both 0.532 and 0.468 are preset weights.

[0029] Furthermore, the method for obtaining the attention of the product according to the extraction of live broadcast dialogue keywords is as follows:

[0030] Obtain the conversations in n live broadcasts, where the conversations are the messages sent by the purchaser to the merchant through the live broadcast platform during the live broadcast;

[0031] Obtain the basic information of the product, where the basic information includes the product name and the mention popularity; at the start of this live broadcast, the mention popularity is zero;

[0032] Retrieve keywords from the conversations, where the keywords correspond to the names of historical products and products on sale;

[0033] Match the keywords with the basic information of historical products and products on sale respectively. Each time keyword matching occurs for product j, the corresponding mention popularity is incremented by one to obtain the attention of the product.

[0034] Furthermore, it also includes obtaining the mention frequency:

[0035] Arbitrarily select one historical product;

[0036] Retrieve the conversations in n live broadcasts, retrieve keywords from the conversations, and obtain the time from the present when each successful match occurs between the keywords and the basic information of historical products, marked as the mention time Yjk, where k = 1, 2, 3, …, f, and f is a positive integer indicating that product j is mentioned f times;

[0037] Mention frequency

[0038] Combine the attention and mention frequency of historical products to obtain the corresponding restocking value Hj of the product, where Hj = attention / (1 + mention frequency).

[0039] Furthermore, the way historical products are restocked is as follows:

[0040] Calculate the recommended value of the product, where recommended value = restocking value + selling value;

[0041] Sort the recommended values of the products in descending order;

[0042] During this live broadcast, calculate and sort the recommended values of the products every preset time T;

[0043] Restock the historical products whose recommended values have appeared in the top five for three consecutive times. If the historical product to be restocked is the same as the product on sale in this live broadcast, increase the quantity of the product on sale by γ times;

[0044] Among them,

[0045]

[0046] Furthermore, the way products on sale are taken off the shelf is as follows:

[0047] Calculate the devaluation value of the product, where devaluation value = 0.315 * attention of the product on sale + 0.685 * selling value;

[0048] Sort the devaluation values of the products in descending order;

[0049] During this live broadcast, calculate and sort the devaluation values of the products every preset time T;

[0050] Take off the shelf the products on sale whose recommended values have appeared in the top five for three consecutive times;

[0051] The quantity taken off the shelf is: 0.6 times the remaining quantity of the product;

[0052] Among them, 0.315, 0.685, and 0.6 are all preset weight values.

[0053] The present invention has the following beneficial effects:

[0054] The present invention obtains the sales records of the goods sold by the merchant on the live streaming platform, marks the remaining goods as historical goods after removing redundant goods; obtains the information of the goods sold by the merchant in this session and marks the corresponding goods as on-sale goods; obtains the sell-out ratio and selling profit of the goods according to the information of the goods sold in this session, and calculates the selling value according to the sell-out ratio and selling profit; obtains the attention degree and restocking value of the goods according to the extraction of the live conversation keywords; combines the attention degree, restocking value and selling value to put the historical goods on the shelf or take the on-sale goods off the shelf, achieving the purpose of being able to timely remind the merchant to adjust the status of the goods and ensure the profit obtained by the creator according to the analysis of various interaction information of the user on the live streaming platform.

[0055] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0057] Figure 1 It is a schematic structural diagram of the method for automatically putting virtual goods on and off the shelf based on a network live streaming platform according to the present invention;

[0058] Figure 2 It is a schematic diagram for obtaining the attention degree in the present invention;

[0059] Figure 3 It is a schematic diagram for obtaining the restocking value in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0061] Please refer to Figures 1 - 3 As shown, the present invention is a method for automatically putting virtual goods on and off the shelf based on a network live streaming platform, including the following steps:

[0062] Step 1: Obtain the sales records of the goods sold by the merchant on the live streaming platform, and mark the remaining goods as historical goods after removing redundant goods;

[0063] Step 2: Obtain the product information sold by the merchant in this session, and mark the corresponding products as on-sale products;

[0064] Step 3: Obtain the sell-out ratio and selling profit of the products according to the product information sold in this session, and calculate the selling value based on the sell-out ratio and selling profit;

[0065] Step 4: Conduct past retrieval: Extract according to the live broadcast dialogue keywords to obtain the attention and restocking value of the products;

[0066] Step 5: Combine the attention, restocking value, and selling value to list historical products or take off-shelf on-sale products;

[0067] As an embodiment provided by the present invention, preferably, the method for taking off-shelf of on-sale products is as follows:

[0068] Calculate the consumption value of the product, consumption value = attention of on-sale product * 0.315 + selling value * 0.685;

[0069] Arrange the consumption values of the products in descending order;

[0070] During this live broadcast, calculate the consumption value of the product and sort it every preset time T;

[0071] Take off-shelf the on-sale products whose recommended values appear in the top five for three consecutive times;

[0072] The quantity of products taken off-shelf is: 0.6 times the remaining quantity of the product;

[0073] Among them, 0.315, 0.685, and 0.6 are all preset weight values.

[0074] As an embodiment provided by the present invention, preferably, the product sales record includes the product name and the corresponding unit price, negative review rate, sales volume, listing time, and sold time for each live broadcast. The method for obtaining the historical products is as follows:

[0075] Obtain the listing time, sold time, negative review rate, and profit of the product from the product sales record;

[0076] Calculate the selling duration Stj, where Ssij is the listing time of product j in the i-th live broadcast, Scij is the sold time of product j in the i-th live broadcast, i = 1, 2, 3,..., n; n ≥ 15, that is, the number of selected live broadcasts in this embodiment is greater than or equal to 15 times. Preferably, select the n live broadcasts closest to this live broadcast; Scjmax and Ssjmin are the longest and shortest times experienced from listing to selling of product j in n live broadcasts; j = 1, 2, 3,..., m; m is a positive integer;

[0077] Obtain the recommended resistance value Tzj of product j, $C_j$ is the negative review rate of product $j$ after $n$ live broadcasts, and $L_j$ is the average profit of product $j$ after $n$ live broadcasts;

[0078] Mark the products whose recommended resistance values appear in the last five for three consecutive times as redundant products, and the rest as historical products.

[0079] As an embodiment provided by the present invention, preferably, the method for obtaining the sell - out ratio of a product according to the product information sold in this session is as follows:

[0080] Arbitrarily select a product for sale and mark it as the target product;

[0081] Obtain the total number of pieces, the shelf - time, and the sell - out time of the target product;

[0082] With the help of the sell - out time and the shelf - time, obtain the sales duration of the target product, where the sales duration = sell - out time - shelf - time;

[0083] With the help of the sales duration and the total number of pieces of the target product, obtain the sell - out ratio. Among them, the discount of the target product in this live broadcast is $\beta$. For example, if product $j$ is sold at a 10% discount in this live broadcast, then $\beta = 0.9$;

[0084] If the target product is not sold out in this live broadcast, then the sell - out ratio = shelf - time / number of pieces sold;

[0085] The sales duration and the shelf - time are calculated in minutes.

[0086] As an embodiment provided by the present invention, preferably, the method for obtaining the selling profit of a product according to the product information sold in this session is as follows:

[0087] S1: Calculate the gross profit of the target product, where the gross profit = selling unit price - cost;

[0088] S2: The selling profit = number of pieces sold * gross profit - sell - out ratio * number of pieces sold * $\alpha$, where $\alpha$ is the live - broadcast cost per minute.

[0089] As an embodiment provided by the present invention, preferably, calculate the selling value according to the sell - out ratio and the selling profit, where the selling value = 0.532 * sell - out ratio+0.468 * selling profit, and 0.532 and 0.468 are both preset weight values.

[0090] As Figure 2 shown, as an embodiment provided by the present invention, preferably, the method for obtaining the attention of a product according to the extraction of live - broadcast dialogue keywords is as follows:

[0091] Obtain the conversations in $n$ live broadcasts, where the conversations are the messages sent by the purchaser to the merchant through the live - broadcast platform during the live - broadcast viewing;

[0092] Obtain the basic information of the commodity, where the basic information includes the commodity name and the mention popularity; at the start of this live broadcast, the mention popularity is zero;

[0093] Retrieve keywords from the conversation, where the keywords correspond to the names of historical commodities and on-sale commodities;

[0094] Match the keywords with the basic information of historical commodities and on-sale commodities respectively. For each time a keyword is matched to commodity j, the corresponding mention popularity is incremented by one to obtain the attention of the commodity.

[0095] As Figure 3 shown, as an embodiment provided by the present invention, preferably, it further includes obtaining the mention time frequency:

[0096] Arbitrarily select a historical commodity;

[0097] Obtain the conversations in n live broadcasts, retrieve keywords from the conversations, and obtain the time from the current time when each successful match between the keywords and the basic information of the historical commodity occurs, which is marked as the mention time Yjk, where k = 1, 2, 3,..., f, and f is a positive integer indicating that commodity j has been mentioned f times;

[0098] Mention time frequency

[0099] Combine the attention of the historical commodity and the mention time frequency to obtain the corresponding return value Hj of the commodity, where Hj = attention / (1 + mention time frequency).

[0100] As an embodiment provided by the present invention, preferably, the way of putting historical commodities on the shelf is:

[0101] Calculate the corresponding recommendation value of the commodity, where the recommendation value = return value + selling value;

[0102] Arrange the corresponding recommendation values of the commodities in descending order;

[0103] In this live broadcast, calculate and sort the recommendation values of the commodities every preset time T;

[0104] Put on the shelf the historical commodities whose recommendation values have appeared in the top five for 3 consecutive times. If the historical commodities put on the shelf are the same as the on-sale commodities in this live broadcast, then increase the quantity of the on-sale commodities by γ times;

[0105] Among them,

[0106]

[0107] The method for automatically putting virtual goods on and off the shelves based on an online live streaming platform obtains the sales records of the goods sold by a merchant on the live streaming platform, marks the remaining goods as historical goods after eliminating redundant goods; obtains the information of the goods sold by the merchant in this session and marks the corresponding goods as on-sale goods; obtains the sell-out ratio and selling profit of the goods according to the information of the goods sold in this session, and calculates the selling value according to the sell-out ratio and selling profit; obtains the attention degree and restocking value of the goods according to the extraction of live chat keywords; combines the attention degree, restocking value, and selling value to put historical goods on the shelves or take on-sale goods off the shelves, achieving the purpose of being able to timely analyze various interaction information of users on the live streaming platform, remind the merchant to adjust the status of the goods, and ensure the profit obtained by the creator.

[0108] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0109] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not elaborate on all details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. This specification selects and specifically describes these embodiments to better explain the principle and practical application of the present invention, so that those skilled in the relevant technical field can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A method for automatically putting virtual goods on and off shelves based on an online live streaming platform, characterized in that, it includes the following steps: Step 1: Obtain the sales records of the goods sold by the merchant on the live streaming platform, and mark the remaining goods as historical goods after eliminating redundant goods; Step 2: Obtain the information of the goods sold by the merchant in this session, and mark the corresponding goods as on-sale goods; Step 3: Obtain the sell-out ratio and selling profit of the goods according to the information of the goods sold in this session, and calculate the selling value according to the sell-out ratio and selling profit; Step 4: Conduct past retrieval: Extract keywords from the live streaming conversation to obtain the attention and restocking value of the goods; Step 5: Combine the attention, restocking value, and selling value to put historical goods on the shelf or take off-sale goods off the shelf; The goods sales record includes the goods name and the unit price, negative review rate, sales volume, shelf time, and sold time corresponding to each live stream. The method for obtaining the historical goods is: Obtain the shelf time, sold time, negative review rate, and profit of the goods from the goods sales record; Calculate the selling time Stj, Where, Ssij is the time when product j is put on the shelves in the i-th live broadcast, Scij is the time when product j is sold in the i-th live broadcast, i = 1, 2, 3, ..., n; n ≥ 15; Scjmax and Ssjmin are the longest and shortest time from the time product j is put on the shelves to the time it is sold in n live broadcasts; j = 1, 2, 3, ..., m; m is a positive integer; Obtain the recommended resistance value Tzj of product j, Cj is the negative review rate of product j after n live broadcasts, and Lj is the average profit of product j after n live broadcasts; Mark the goods whose recommended resistance value appears in the last five for three consecutive times as redundant goods, and the rest as historical goods; The method for obtaining the sell-out ratio of the goods according to the information of the goods sold in this session is: Select any on-sale good and mark it as the target good; Obtain the total number of pieces, shelf time, and sell-out time of the target good; With the help of the sell-out time and shelf time, obtain the sales duration of the target good, and the sales duration = sell-out time - shelf time; The sell-out ratio is obtained by means of the sales duration and total number of pieces of the target product. Among them, the discount of the target product in this live broadcast is β. If the target good is not sold out in this live stream, then the sell-out ratio = shelf duration / number of pieces sold; The sales duration and shelf duration are calculated in minutes; The method for obtaining the selling profit of the goods according to the information of the goods sold in this session is: S1: Calculate the gross profit of the target good, and the gross profit = selling unit price - cost; S2: The selling profit = number of pieces sold * gross profit - sell-out ratio * number of pieces sold * α, where α is the live streaming cost per minute; Calculate the selling value according to the sell-out ratio and selling profit, and the selling value = 0.532 * sell-out ratio + 0.468 * selling profit, where 0.532 and 0.468 are both preset weight values; The method for obtaining the attention of the goods by extracting keywords from the live streaming conversation is: Obtain the conversations in n live streams, and the conversations are the messages sent by the buyer to the merchant through the live streaming platform during the live stream viewing; Obtain the basic information of the goods, and the basic information includes the goods name and mention popularity; at the start of this live stream, the mention popularity is zero; Retrieve keywords from the conversations, and the keywords correspond to the names of historical goods and on-sale goods; Match the keywords with the basic information of historical goods and on-sale goods respectively. Each time the keyword is matched with the goods j, the corresponding mention popularity is incremented by one to obtain the attention of the goods; It also includes the acquisition of mention frequency: Select any historical good; Obtain the conversations in n live streams, retrieve keywords from the conversations, and obtain the time from the current time when each successful match between the keyword and the basic information of the historical good occurs, and mark it as the mention time Yjk, k = 1, 2, 3,..., f, where f is a positive integer representing that the good j is mentioned f times; When referring to time-frequency Obtain the corresponding restocking value Hj of the product by combining the attention and the frequency of mention of historical products. Hj = attention / (1 + frequency of mention).

2. The method for automatically putting virtual products on and off the shelves based on an online live streaming platform according to claim 1, characterized in that, The way to put historical products on the shelves is as follows: Calculate the corresponding recommendation value of the product. Recommendation value = restocking value + selling value; Arrange the recommendation values of the products in descending order; In this live stream, calculate and sort the recommendation values of the products every preset time T; Put on the shelves the historical products whose recommendation values have appeared in the top five continuously for 3 times. If the historical product put on the shelves is the same as the product on sale in this live stream, increase the quantity of the product on sale by γ times; wherein, 3. The method for automatically putting virtual products on and off the shelves based on an online live streaming platform according to claim 1, characterized in that, The way to take off the products on sale is as follows: Calculate the devaluation value of the product. Devaluation value = attention of the product on sale * 0.315 + selling value * 0.685; Arrange the devaluation values of the products in descending order; In this live stream, calculate and sort the devaluation values of the products every preset time T; Take off the products on sale whose recommendation values have appeared in the top five continuously for 3 times; The quantity taken off is: 0.6 times the remaining quantity of the product; wherein, 0.315, 0.685, and 0.6 are all preset weight values.

Images