Reverse scene-based search method and device

Abstract

The embodiment of the invention discloses a search method and device based on a reverse scene. The method comprises the following steps: based on data stored in a constructed search engine, retrieving commodity information matched with the name of a target reverse commodity, and marking the attribute of the target reverse commodity based on the matched commodity information; and quantifying the attribute of the target reverse commodity, and correcting the preset price of the target reverse commodity based on a quantification result to price the target reverse commodity. According to the method, the retrieval efficiency is high, the attributes of the commodities can be quantified, the same commodity has the same price, pricing is performed under the guidance of the quantified attributes, the price is objective, manual operation is not needed, and the labor cost can be saved.

Description

Technical Field

[0001] This application relates to the field of computers, specifically to the field of the Internet, and in particular to search methods and apparatus based on reverse scenarios. Background Technology

[0002] In the reverse supply chain of secondhand goods, after secondhand goods are received into the warehouse, they need to be evaluated, priced, and sold. Often, secondhand goods suffer from significant information loss, making accurate evaluation difficult. Therefore, it is necessary to manually query the product's price, category, and other attribute information on e-commerce platforms and use experience to assess the product's attributes to price the secondhand goods.

[0003] However, manual inquiry is inefficient, and manual evaluation cannot quantify the attributes of a product, which may result in the same product having different prices at different times. Manual pricing lacks quantitative guidance, is highly subjective, and is difficult to make precise considerations. Manual evaluation and calculation during pricing are costly in terms of manpower. Summary of the Invention

[0004] The embodiments of this application provide a search method and apparatus based on reverse scenarios.

[0005] In a first aspect, embodiments of this application provide a search method based on a reverse scenario, comprising: retrieving product information matching the name of a target reverse product based on data stored in a constructed search engine, and labeling the attributes of the target reverse product based on the matching product information; and quantifying the attributes of the target reverse product and adjusting the preset price of the target reverse product based on the quantification results to price the target reverse product.

[0006] In one embodiment of this application, the method further includes: periodically extracting product data from an e-commerce platform and writing it into an original data table, wherein the product data includes information on forward products and reverse products; and importing the data from the original data table into a constructed search engine to update the data stored in the constructed search engine.

[0007] In one embodiment of this application, importing data from the original data table into the constructed search engine includes: creating a temporary data table containing fields from the constructed search engine; importing product data from the original data table into the temporary data table; and batch importing product data from the temporary data table into the constructed search engine.

[0008] In one embodiment of this application, the search engine is constructed through the following steps: configuring the parameters of the distributed search engine based on the magnitude of the product data of the e-commerce platform; and constructing a distributed search engine with a cluster node number adapted to the magnitude of the product data based on the configured parameters.

[0009] In one embodiment of this application, based on data stored in a constructed search engine, retrieving product information that matches the name of a target reverse product and labeling the attributes of the target reverse product based on the matching product information includes: matching the name of the target reverse product with data stored in the constructed search engine to recall product information that matches the name of the target reverse product; sorting the recalled product information in descending order of time; and labeling the attributes of the target reverse product based on a preset number of product information at the top of the sort, wherein the attributes of the target reverse product include the product category.

[0010] In one embodiment of this application, quantifying the attributes of the target reverse product and adjusting the preset price of the target reverse product based on the quantification result to price the target reverse product includes: quantifying the attributes of the target reverse product; determining optimization parameters that match the sales method based on the quantification result; and adjusting the preset price of the target reverse product based on the optimization parameters and pricing the target reverse product at the adjusted price.

[0011] In one embodiment of this application, determining optimization parameters matching the sales method based on the quantification results includes: determining the auction success probability and auction failure probability matching the auction method based on the quantification results; and adjusting the preset price of the target reverse product based on the optimization parameters and pricing the target reverse product at the adjusted price, including: adjusting the minimum starting price of the target reverse product based on the auction success probability and auction failure probability to obtain the starting price of the target reverse product.

[0012] Secondly, embodiments of this application provide a search device based on a reverse scenario, comprising: an attribute labeling module configured to retrieve product information matching the name of a target reverse product based on data stored in a constructed search engine, and label the attributes of the target reverse product based on the matching product information; and a product pricing module configured to quantify the attributes of the target reverse product and adjust the preset price of the target reverse product based on the quantification result to price the target reverse product.

[0013] In one embodiment of this application, the apparatus further includes: a data extraction module configured to periodically extract product data from an e-commerce platform and write it into an original data table, wherein the product data includes information on forward products and information on reverse products; and a data import module configured to import the data in the original data table into a constructed search engine to update the data stored in the constructed search engine.

[0014] In one embodiment of this application, the data import module includes: a data import unit configured to create a temporary data table containing fields from the constructed search engine; import product data from the original data table into the temporary data table; and batch import product data from the temporary data table into the constructed search engine.

[0015] In one embodiment of this application, the apparatus further includes: a parameter configuration module configured to configure parameter indicators of a distributed search engine based on the magnitude of the product data of the e-commerce platform; and a search engine construction module configured to construct a distributed search engine with a cluster node number adapted to the magnitude of the product data based on the configured parameter indicators.

[0016] In one embodiment of this application, the attribute labeling module includes: a recall unit configured to match the name of the target reverse product with data stored in a constructed search engine, and recall product information that matches the name of the target reverse product; a sorting unit configured to sort the recalled product information in descending order of time; and a labeling unit configured to label the attributes of the target reverse product based on a preset number of product information that are ranked first, wherein the attributes of the target reverse product include the product category.

[0017] In one embodiment of this application, the product pricing module includes: a quantification unit configured to quantify the attributes of the target reverse product; a parameter determination unit configured to determine optimization parameters matching the sales method based on the quantification result; and an optimization unit configured to correct the original price of the target reverse product based on the optimization parameters and price the target reverse product at the corrected price.

[0018] In one embodiment of this application, the parameter determination unit is configured to determine the auction success probability and auction failure probability that match the auction method based on the quantification result; the optimization unit is configured to adjust the minimum starting price of the target reverse commodity based on the auction success probability and auction failure probability to obtain the starting price of the target reverse commodity.

[0019] Thirdly, embodiments of this application provide a search engine cluster including multiple nodes, each node including: one or more processors; and a memory for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors of the multiple nodes to implement the method described in the first method.

[0020] Fourthly, embodiments of this application provide a computer-readable medium having a computer program stored thereon that, when executed by a processor, implements the method described in the first aspect.

[0021] The reverse scenario-based search method and apparatus provided in this application retrieve product information by constructing a dedicated search engine for reverse scenarios, and annotates the target reverse products with attributes based on the retrieved product information, quantifies the labeled attributes of the target reverse products, and guides the pricing of the target reverse products based on the quantified attribute results. The retrieval efficiency is high, the product attributes can be quantified, so that the same product has the same price, and the pricing is objective under the guidance of quantified attributes. No manual operation is required, which can save labor costs. Attached Figure Description

[0022] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0023] Figure 1 This is a diagram illustrating the process of secondhand goods going from warehousing to sale in a reverse scenario;

[0024] Figure 2 This is an exemplary system architecture diagram of a search method based on reverse scenarios suitable for implementing the embodiments of this application;

[0025] Figure 3 This is a flowchart of a search method based on a reverse scenario according to an embodiment of this application;

[0026] Figure 4 This is a flowchart illustrating the method for labeling product attributes according to the embodiments of this application;

[0027] Figure 5 This is a flowchart illustrating the method for product pricing according to the embodiments of this application;

[0028] Figure 6 This is a schematic diagram of an embodiment of a method for pricing goods according to the implementation of this application;

[0029] Figure 7 This is a flowchart illustrating the data update method according to the embodiments of this application;

[0030] Figure 8 This is a flowchart illustrating the method for constructing a search engine according to an embodiment of this application;

[0031] Figure 9 This is a schematic diagram of a search device based on a reverse scenario according to an embodiment of this application;

[0032] Figure 10 This is a block diagram of an electronic device used to implement the reverse scenario-based search method of the embodiments of this application. Detailed Implementation

[0033] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0034] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0035] Product supply encompasses both forward and reverse scenarios. In the forward scenario, goods are delivered to customers through the forward supply chain. Existing e-commerce platforms primarily handle forward supply, thus their product searches are concentrated in this area. In the reverse scenario, goods are delivered to customers through the reverse supply chain. The sale of secondhand goods falls under the reverse supply chain category. Existing e-commerce platforms show less focus on searching for reverse-sector products and often lack the capability to support such searches. Furthermore, given the large scale and numerous cluster nodes of existing e-commerce search engines, directly applying them to the reverse sector would result in resource waste.

[0036] Figure 1 This illustrates the process of secondhand goods from warehousing to sale in a reverse scenario. For example... Figure 1 As shown, Figure 1 The above is the existing business process. After second-hand goods are put into storage, they need to be evaluated, priced, and sold. Since second-hand goods often have serious information loss, it is difficult to make an accurate evaluation. Therefore, after second-hand goods are put into storage, the price, category and other attribute information of the goods need to be manually queried on the e-commerce platform first. Then, the attributes of the goods are evaluated by the staff based on experience to price the second-hand goods. Only after that, the priced second-hand goods are sold through offline stores or online platforms.

[0037] However, the existing business processes have the following problems:

[0038] 1. Manual querying is inefficient, typically yielding only 1-2 results per minute;

[0039] 2. Manual evaluation cannot quantify the attributes of a product, which may result in the same product having different prices at different times;

[0040] 3. Manual pricing lacks quantitative guidance, is highly subjective, and makes it difficult to achieve refined consideration;

[0041] 4. Pricing requires manual evaluation and calculation, resulting in high labor costs.

[0042] To address the aforementioned issues, the embodiments of this application provide a search method based on reverse scenarios.

[0043] Figure 2 An exemplary system architecture 100 for a reverse scenario-based search method suitable for implementing embodiments of this application is shown. Figure 2 As shown, the system architecture 100 may include multiple nodes 101, 102, 103, 104... These nodes 101, 102, 103, 104... communicate with each other to form a distributed search engine cluster. Each node 101, 102, 103, 104... can be hardware or software. When a node is hardware, it can be implemented as a single physical machine. When a node is software, it can be implemented as a single virtual machine or container. The implementation methods in this application do not limit this.

[0044] It should be noted that, Figure 2 For ease of explanation, only 4 nodes in the distributed search engine cluster are shown. However, the number of nodes in the distributed search engine cluster of this application is not limited to 4. For example, the number of nodes in the distributed search engine cluster can also be 5, 10, 20, 40, etc.

[0045] It should be noted that the composition Figure 2 The multiple nodes 101, 102, 103, 104... in the distributed search engine cluster can be nodes of the same type, such as all nodes being physical machines, or they can be nodes of different types, such as nodes including physical machines, virtual machines and containers. The implementer of this application does not limit this.

[0046] Figure 3 The flowchart of a reverse scenario-based search method according to an embodiment of this application is shown. The reverse scenario-based search method provided by the embodiments of this application can be... Figure 2 The search engine cluster in the process is executed. For example... Figure 3 As shown, the method may include the following steps:

[0047] Step 201: Based on the data stored in the constructed search engine, retrieve product information that matches the name of the target reverse product and label the attributes of the target reverse product based on the matching product information.

[0048] In the embodiments of this application, a search engine specifically designed for reverse scenarios can be built by accumulating a large amount of product information to retrieve product information and evaluate pricing.

[0049] Alternatively, a search engine specifically designed for reverse engineering can be built based on an open-source search engine, such as the open-source ElasticSearch search engine. The implementation of this application does not limit the type of open-source search engine used to build the search engine specifically designed for reverse engineering.

[0050] Optionally, a large amount of product information can be accumulated from the data of e-commerce platforms. The large amount of product information accumulated from e-commerce platforms can include both reverse product information and forward product information. The amount of accumulated product information can reach hundreds of millions. The implementation method of this application does not limit the implementation method of accumulating product information from e-commerce platforms.

[0051] Optionally, product information may include product characteristics such as name, price, category, condition, historical sales volume, historical price, seasonal sales trends, and holiday sales. The types of product characteristics included in each piece of product information stored in the search engine can be the same (e.g., the product information only includes general product characteristics) or different (e.g., the product information includes both general and specific product characteristics). The types of product characteristics included in the product information can be set as needed, and the embodiments of this application do not limit this.

[0052] In one optional example, the types of product characteristics included in the product information for new and used goods differ. For example, the product information for new goods may include: product name, price, category, etc., while the product information for used goods may include: product name, category, condition, starting price, and final price, etc. In another optional example, the types of product characteristics included in the product information for different types of goods differ. For example, the product information for food products may include: product name, price, category, and shelf life, while the product information for non-food products may include: product name, price, category, and condition, etc. Here, the condition of the product refers to its degree of wear and tear. The category of the product can have different definitions for different types of goods. For example, for clothing, the product category may be Class A, Class B, and Class C in the clothing safety rating system; for porcelain, the product category may be Grade 1, Grade 2, Grade 3, and Grade 4 in the porcelain quality rating system.

[0053] In the embodiments of this application, after the target reverse-engineered goods are stored in the warehouse, product information can be retrieved using a search engine specifically designed for reverse engineering, based on the name of the target reverse-engineered goods. The search engine searches for product information matching the name of the target reverse-engineered goods from its vast database, and then labels the attributes of the target reverse-engineered goods based on the found matching product information. Here, reverse-engineered goods can refer to recycled second-hand goods, which can include both used and unused goods.

[0054] Optionally, the attributes of a product may include the types of product features included in the product information of all products with the same name, or may only include a portion of the types of product features included in the product information of all products with the same name. The types of product features included in the attributes of different types of products may be the same or different. The types of product features included in the attributes of a product can be set as needed, and the embodiments of this application do not limit this.

[0055] In one optional example, the product information for a new item with the same name includes: product name, price, and category; the product information for a used item with the same name includes: product name, category, condition, starting price, and final price. Product attributes may include: product name, new product price, category, condition, starting price, and final price. Alternatively, in another optional example, the product information for a new item with the same name includes: product name, price, and category; the product information for a used item with the same name includes: product name, category, condition, starting price, and final price. Product attributes may only include: product name and category.

[0056] Step 202: Quantify the attributes of the target reverse product and adjust the preset price of the target reverse product based on the quantification results to price the target reverse product.

[0057] In the embodiments of this application, an optimization model can be preset, the attributes of the target reverse product can be quantified by the optimization model, and the preset price of the target reverse product can be corrected according to the quantified value of the obtained attributes, and the corrected price can be determined as the selling price of the target reverse product.

[0058] Optionally, at least one of the following quantification methods—categorical quantification, ordinal quantification, interval quantification, and ratio quantification—can be used to quantify the attributes of a product. The embodiments of this application do not limit this. Different attributes of a product can be quantified using the same quantification method; for example, the category and quality of a product can both be quantified using categorical quantification. Alternatively, different quantification methods can be used; for example, the category of a product can be quantified using ordinal quantification, and the quality of a product can be quantified using interval quantification. The embodiments of this application do not limit this.

[0059] Optionally, the preset price of the target reverse product can be a price set based on the buyback price. For example, it can be the buyback price of the target reverse product multiplied by a preset value, which can be greater than 1, less than 1, or equal to 1. In one optional example, for a target reverse product sold through auction, the preset price can be the minimum starting bid, and this preset value can be less than 1. In another optional example, for a target reverse product sold at a fixed price, the preset price can be the original selling price, and this preset value can be greater than 1. The optimization model adjusts the preset price differently for different selling methods; for example, the formula used by the optimization model to adjust the preset price differs between selling through auction and selling at a fixed price.

[0060] like Figure 1 As shown, Figure 1 The following is a pricing process for the search method using the implementation method of this application. After the second-hand goods are put into storage, a large amount of product information is first searched through a search engine specifically designed for reverse scenarios. Based on the search results, the price, category and other attribute information of the second-hand goods are labeled. Then, the attributes of the labeled goods are quantified, and the pricing of the second-hand goods is guided by the quantified values ​​of the attributes. After that, the priced second-hand goods are sold through offline stores or online platforms.

[0061] The reverse scenario-based search method provided in this application retrieves product information by constructing a dedicated search engine for reverse scenarios. It then labels the target reverse product with attributes based on the retrieved product information, quantifies the labeled attributes, and guides the pricing of the target reverse product based on the quantified attribute results. This method boasts high retrieval efficiency, reaching 1000 results per minute, and quantifies product attributes, ensuring that the same product has the same price. Pricing is guided by quantified attributes, resulting in objective pricing. It requires no manual intervention, saving labor costs, and can be applied in offline stores or online platforms.

[0062] It should be understood that, in Figure 3The steps shown are not exclusive; other steps may be performed before, after, or between any of the steps shown. Furthermore, some steps shown may be performed simultaneously, or they may be performed in a sequence different from the steps shown. Figure 3 The execution is performed in the order shown.

[0063] Figure 4 The flowchart illustrating the method for labeling product attributes according to an embodiment of this application is shown. For example... Figure 4 As shown, retrieving product information that matches the name of the target reverse product based on the data stored in the constructed search engine and labeling the attributes of the target reverse product based on the matching product information may include the following steps:

[0064] Step 301: Match the name of the target reverse product with the data stored in the constructed search engine to recall product information that matches the name of the target reverse product.

[0065] In some optional embodiments of this application, the search engine specifically designed for reverse engineering scenarios can be associated with the warehousing registration information of reverse-engineered goods. The search engine can retrieve product information from stored data in response to the addition of the target reverse-engineered product name in the warehousing registration information. Alternatively, the search engine can also respond to manual operations, retrieving the target reverse-engineered product name from the warehousing registration information, such as retrieving the name of target reverse-engineered goods that entered the warehouse within a certain period, or retrieving the name of target reverse-engineered goods that were manually marked, and then retrieving product information from stored data. The embodiments of this application do not limit the implementation method of the search engine specifically designed for reverse engineering scenarios retrieving information based on the name of the target reverse-engineered product.

[0066] Optionally, the name of the product in the recalled product information may be exactly the same as the name of the target reverse product, or it may include the name of the target reverse product. That is, the name of the target reverse product is part of the name of the product in the recalled product information. The implementation of this application does not limit this.

[0067] Optionally, the search engine specifically designed for reverse engineering scenarios can, in addition to using the name of the target reverse product as a keyword for retrieval, also use other product characteristics of the target reverse product along with the name of the target reverse product as keywords in the stored data for retrieving product information. For example, the name and category of the target reverse product can be used as keywords for retrieving product information. The implementation method of this application does not limit this.

[0068] Step 302: Sort the recalled product information in descending order of time.

[0069] In some optional embodiments of this application, the search engine specifically designed for reverse scenarios can sort the recalled product information by time after recalling the product information, in descending order of time, placing the product information most recent to the current time at the front and the product information furthest from the current time at the back. The recalled product information can include both second-hand product information and new product information.

[0070] Step 303: Label the attributes of the target reverse product based on the preset number of product information at the top of the sort, wherein the attributes of the target reverse product include the product category.

[0071] In some optional embodiments of this application, since the same product has different historical prices, product information more recent than the current time is more instructive for pricing the target reverse product. Therefore, when labeling the attributes of the target reverse product, the search engine specifically designed for reverse scenarios can select only a preset number of product information items ranked at the top to label the attributes of the target reverse product. The preset number can be set based on experience; for example, the preset number can be 5, and the search engine specifically designed for reverse scenarios can select only the top 5 product information items to label the attributes of the target reverse product.

[0072] Optionally, the preset quantity can be different for different types of goods. For example, for fast-moving consumer goods, whose prices change quickly, the preset quantity can be as large as possible, while for slow-moving consumer goods, whose prices change slowly, the preset quantity can be as small as possible.

[0073] The implementation method of this application uses a search engine specifically designed for reverse engineering scenarios to annotate the attributes of the target reverse product based on the retrieval of product information. This can ultimately provide detailed information about the target reverse product, enabling information tracing of the target reverse product, refining the target reverse product profile, and providing a basis for guiding the pricing of the target reverse product.

[0074] Figure 5 The process of pricing goods according to an embodiment of this application is illustrated. For example... Figure 5 As shown, quantifying the attributes of the target reverse product and adjusting the preset price of the target reverse product based on the quantification results to price the target reverse product may include the following steps:

[0075] Step 401: Quantify the attributes of the target reverse product.

[0076] In some optional embodiments of this application, the attributes labeled on the target reverse product can be quantified by an optimization model. Optionally, all attributes labeled on the target reverse product can be quantified, or only a portion of the attributes labeled on the target reverse product can be quantified. The embodiments of this application do not limit this.

[0077] Optionally, when quantifying only a subset of the attributes labeled on the target reverse product, the attributes to be quantified can differ for different types of products. For example, for food products, only the price, category, and shelf life can be quantified, while for non-food products, only the price, category, and quality can be quantified. When quantifying only a subset of the attributes labeled on the target reverse product, the attributes that need to be quantified can be set as needed, and the implementation method of this application does not limit this.

[0078] Step 402: Determine the optimal parameters that match the sales method based on the quantitative results.

[0079] In some optional embodiments of this application, after quantifying the attributes of the target reverse product, the optimization model can determine the corresponding optimization parameters based on the quantified values ​​of the target reverse product's attributes according to the sales method of the target reverse product. Optionally, when the target reverse product has multiple quantified attributes, a weight can be assigned to each attribute according to the importance of the multiple attributes in the product pricing, and the optimization parameters can be determined by weighting the quantified values ​​of each attribute.

[0080] Optionally, the sales method of the target reverse product can be determined based on the sales method of offline stores or online platforms, or it can be determined based on the type of the target reverse product. This application's implementation does not limit this. Optionally, since the formulas used for price adjustment differ for different sales methods, and the corresponding optimization parameters in the formulas also differ, the optimization model determines the optimization parameters based on the sales method of the target reverse product, ensuring that the determined optimization parameters match the sales method of the target reverse product. For example, if the target reverse product is sold through auction, the optimization model determines the auction success probability and auction failure probability, matching the auction method, as optimization parameters based on the quantified results.

[0081] Step 403: Adjust the preset price of the target reverse product based on the optimization parameters and price the target reverse product with the adjusted price.

[0082] In some optional embodiments of this application, after determining the optimization parameters that match the sales method, the optimization model can use the optimization parameters to correct the preset price to obtain the corrected price. The corrected price is used as the selling price of the target reverse product, which can realize the quantitative guidance of product attributes on product pricing. Since the characteristics of the product are carefully considered in the pricing process, the pricing of the product can be more reasonable.

[0083] Optionally, the preset price differs for different sales methods, and the corrected price obtained by adjusting the preset price using optimization parameters will also differ. For example, if the target reverse product is sold through auction, the optimization model can adjust the minimum starting price of the target reverse product based on the auction success probability and the auction failure probability to obtain the starting price of the target reverse product.

[0084] In an optional example, such as Figure 6 As shown, Figure 6 This diagram illustrates an embodiment of a method for optimizing commodity pricing according to the present application. Second-hand goods are sold through auction on an online platform. The minimum starting bid r for the second-hand goods can be determined by multiplying the buyback price of the second-hand goods by a value less than 1. m The attributes of secondhand goods can include condition, price as new, starting bid, and final bid. The probability of a successful auction (p) and the probability of an unsuccessful auction (punsuccessful) can be determined based on the quantified values ​​of these attributes. Then, the minimum starting bid (r) can be adjusted according to Formula 1 in the optimization model. m The optimized starting price r is obtained through optimization. Formula 1 is as follows:

[0085]

[0086]

[0087]

[0088] P(r)≥r m

[0089] 0≤p≤1

[0090]

[0091] Where R(r) represents revenue, and the formula aims to maximize the seller's revenue in its pricing decision; p represents the probability of a successful auction; ~p represents the probability of an unsuccessful auction; P(r) represents the final auction price; i represents the percentage increase in bids; r represents the starting price; r m The minimum starting bid is represented by k*; the number of bid increments is represented by k*; Cd represents costs such as product loss; Cc(r) represents platform fees; σ represents the minimum bid increment for each bidder; V NThis represents the upper limit of the customer's estimated price. Formula 1 also considers the probability of a sale, which helps to minimize potential losses when the discount is lowered again after a product has not sold within a certain period, thus maximizing the seller's profits.

[0092] Figure 7 The flowchart illustrating a method for updating data according to an embodiment of this application is shown. For example... Figure 7 As shown, the search method based on reverse scenarios in the embodiments of this application may further include the following steps:

[0093] Step 501: Periodically extract product data from e-commerce platforms and write it into the original data table. The product data includes information on forward products and reverse products.

[0094] In some optional embodiments of this application, product data from e-commerce platforms can be extracted periodically and written into an original data table. Optionally, the original data table can be a HIVE table, a data warehouse tool based on Hadoop used for data extraction, transformation, and loading. HIVE is a mechanism for storing, querying, and analyzing large-scale data stored in Hadoop. Optionally, product data from e-commerce platforms can be extracted and written into the original data table daily.

[0095] Step 502: Import the data from the original data table into the constructed search engine to update the data stored in the constructed search engine.

[0096] In some optional embodiments of this application, product data from the original data table can be imported into a search engine specifically designed for reverse engineering scenarios to update the data stored in the search engine. Optionally, when product data is stored in the search engine, the original data table can be updated incrementally; when no product data is stored in the search engine, the original data table can be updated with a full update.

[0097] Optionally, importing data from the original data table into the constructed search engine may include the following steps: creating a temporary data table containing the fields of the constructed search engine; importing product data from the original data table into the temporary data table; and batch importing the product data from the temporary data table into the constructed search engine. After importing the product data from the original data table into the temporary data table, the necessary environment files for the constructed search engine can be added to the constructed search engine, allowing data from the temporary data table to be batch imported into the constructed search engine. By creating a temporary table to import data from the original data table into the constructed search engine, the data import time can be shortened, and the efficiency of data import can be improved.

[0098] In an optional example, the original data table uses a HIVE table, and the search engine specifically designed for reverse engineering is built on the open-source Elasticsearch search engine. When importing data from the HIVE table into the Elasticsearch search engine, a temporary table containing fields from Elasticsearch can be created first, denoted as `app.app_mtg_gmd_hive_to_es`. Then, the product data to be imported from the HIVE table into the temporary table `app.app_mtg_gmd_hive_to_es`. Afterward, the necessary JAR files for the Elasticsearch search engine are added, and the data from the temporary table `app.app_mtg_gmd_hive_to_es` is imported in batches into the Elasticsearch search engine. Using this method, the data import efficiency can be improved to 400 million records per hour, compared to 30 million records per hour using the Elasticsearch library in Python, representing a 1300% improvement in efficiency.

[0099] Figure 8 The process of constructing a search engine according to an embodiment of this application is illustrated. For example... Figure 8 As shown, the search engine of the embodiments of this application is constructed through the following steps:

[0100] Step 601: Configure the parameter metrics of the distributed search engine based on the magnitude of the product data of the e-commerce platform.

[0101] In some optional embodiments of this application, the parameters of the distributed search engine can be configured according to the amount of product data that needs to be stored in the distributed search engine. In one optional example, the distributed search engine uses the open-source Elasticsearch search engine, and the amount of product data extracted from the e-commerce platform is 600 million+. Based on the amount of 600 million+, the Elasticsearch search engine can be configured with parameters such as the ES version number of the distributed nodes, the number of nodes, the number of CPU cores per node, the memory capacity, and the hard disk capacity, etc. The configured parameters of the Elasticsearch search engine are shown in Table 1.

[0102] Table 1

[0103] ES version number Number of nodes CPU cores Memory capacity Hard drive capacity 7.10.2_basic 40 2 16G 20G

[0104] Step 602: Construct a distributed search engine based on the configured parameter indicators, with the number of cluster nodes matching the magnitude of the product data.

[0105] In some optional embodiments of this application, after configuring the parameters of the distributed search engine, a distributed search engine cluster can be constructed based on the configured parameters. The number of nodes in this distributed search engine cluster can be matched to the magnitude of the product data, thus avoiding both resource waste due to an excessive number of nodes and excessively low requests per second (QPS) due to an insufficient number of nodes. In one optional example, the distributed search engine uses the open-source ElasticSearch search engine, extracting over 600 million product data points from an e-commerce platform. The data metrics of the ElasticSearch search engine specifically built for reverse engineering are shown in Table 2.

[0106] Table 2

[0107]

[0108] In an optional example, an A / B test is conducted, with the experimental group using the reverse scenario-based search method of the present application to price secondhand goods, and the control group using existing methods (such as...). Figure 1 The existing business process above is used to price second-hand goods, and the experimental results are shown in Table 3.

[0109] Table 3

[0110]

[0111] In Table 3, the price before the price change represents the original price of the secondhand goods, and the price after the price change represents the price after the repricing. In the Northeast region of a certain online platform, before the price change, the average daily profit of the experimental group was 778.19, and the average daily profit of the control group was 748.27. After the price change, compared to before the price change, the experimental group's profit increased by 10.30% compared to the control group. In the East China region of the same online platform, before the price change, the average daily profit of the experimental group was 833.52, and the average daily profit of the control group was 818.27. After the price change, compared to before the price change, the experimental group's profit increased by 15.66% compared to the control group.

[0112] Figure 9 A schematic diagram of a reverse-scenario-based search apparatus according to an embodiment of this application is shown. Figure 9 As shown, the reverse scenario-based search device 700 of this application may include: an attribute labeling module 701 and a product pricing module 702. Wherein,

[0113] The attribute labeling module 701 is configured to retrieve product information matching the name of the target reverse product based on data stored in the constructed search engine, and to label the attributes of the target reverse product based on the matching product information; and

[0114] The product pricing module 702 is configured to quantify the attributes of the target reverse product and optimize the preset price of the target reverse product based on the quantification results to price the target reverse product.

[0115] In some alternative embodiments of this application, the device 700 may further include:

[0116] The data extraction module is configured to periodically extract product data from the e-commerce platform and write it into a raw data table. This product data includes information on both forward and reverse-product items.

[0117] The data import module is configured to import data from the original data table into the constructed search engine to update the data stored in the constructed search engine.

[0118] In some optional embodiments of this application, the data import module may further include:

[0119] The data import unit is configured to create a temporary data table containing fields from the constructed search engine; import product data from the original data table into the temporary data table; and import product data from the original data table in batches into the constructed search engine based on the temporary data table.

[0120] In some alternative embodiments of this application, the device 700 may further include:

[0121] The parameter configuration module is configured to configure the parameter metrics of the distributed search engine based on the order of magnitude of the product data from the e-commerce platform; and

[0122] The search engine building module is configured to build a distributed search engine based on configured parameter indicators, with the number of cluster nodes matching the order of magnitude of the product data.

[0123] In some optional embodiments of this application, the attribute annotation module 701 may include:

[0124] The recall unit is configured to match the name of the target reverse product with the data stored in the constructed search engine, and recall product information that matches the name of the target reverse product.

[0125] The sorting unit is configured to sort the recalled product information in descending order of time; and

[0126] The labeling unit is configured to label the attributes of the target reverse product based on a preset number of product information that are ranked first, wherein the attributes of the target reverse product include the product category.

[0127] In some optional embodiments of this application, the product pricing module 702 may include:

[0128] The quantization unit quantifies the attributes of the target reverse product;

[0129] The parameter determination unit is configured to determine optimized parameters that match the sales method based on the quantization results; and

[0130] The optimization unit is configured to adjust the preset price of the target reverse product based on the optimization parameters and price the target reverse product at the adjusted price.

[0131] In some optional embodiments of this application, the parameter determination unit is configured to determine the auction success probability and auction failure probability that match the auction method based on the quantification results;

[0132] The optimization unit is configured to optimize the minimum starting price of the target reverse commodity based on the auction success probability and the auction failure probability to obtain the starting price of the target reverse commodity.

[0133] It should be understood that the various components and units in the reverse scenario-based search device 700 are related to the reference... Figures 3 to 8 The steps described in the method correspond to those in the previous section. Therefore, the operations and features described above for the search method based on the reverse scenario are also applicable to the device 700 and the modules and units contained therein, and will not be repeated here.

[0134] According to embodiments of this application, this application also provides a search engine cluster and a readable storage medium.

[0135] Figure 10 A block diagram of an electronic device 800 performing a reverse scenario-based search method according to an embodiment of this application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present application described and / or claimed herein.

[0136] like Figure 10 As shown, device 800 includes a processor 801, which can perform various appropriate actions and processes according to a computer program stored in read-only memory (ROM) 802 or a computer program loaded from memory 808 into random access memory (RAM) 803. RAM 803 may also store various programs and data required for the operation of device 800. The processor 801, ROM 802, and RAM 803 are interconnected via bus 804. I / O interface (input / output interface) 805 is also connected to bus 804.

[0137] Multiple components in device 800 are connected to I / O interface 805, including: input unit 806, such as keyboard, mouse, etc.; output unit 807, such as various types of displays, speakers, etc.; memory 808, such as disk, optical disk, etc.; and communication unit 809, such as network card, modem, wireless transceiver, etc. Communication unit 809 allows device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0138] Processor 801 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 801 performs the various methods and processes described above, such as the reverse scenario-based search method. For example, in some embodiments, the reverse scenario-based search method may be implemented as a computer software program tangibly contained in a machine-readable storage medium, such as memory 808. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by processor 801, one or more steps of the reverse scenario-based search method described above may be performed. Alternatively, in other embodiments, processor 801 may be configured to perform the reverse scenario-based search method by any other suitable means (e.g., by means of firmware).

[0139] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0140] The program code used to implement the methods of this application may be written in any combination of one or more programming languages. The program code may be packaged into a computer program product. This program code or computer program product may be provided to the processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor 801, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0141] In the context of this application, a machine-readable storage medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable storage medium can be a machine-readable signal storage medium or a machine-readable storage medium. A machine-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0142] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0143] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0144] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service ecosystem, addressing the shortcomings of traditional physical hosts and VPS (Virtual Private Server, or simply "VPS") services, such as high management difficulty and weak business scalability. A server can also be a server in a distributed system or a server incorporating blockchain technology.

[0145] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this application can be achieved, and this is not limited herein.

[0146] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A search method based on reverse scenarios, comprising: Based on the data stored in the constructed search engine, retrieve product information that matches the name of the target reverse product and label the attributes of the target reverse product based on the matching product information; as well as The attributes of the target reverse product are quantified, and the preset price of the target reverse product is adjusted based on the quantification results to set the price of the target reverse product.

2. The method according to claim 1, wherein, The method further includes: Periodically extract product data from e-commerce platforms and write it into a raw data table. This product data includes information on both forward and reverse-product listings. Import the data from the original data table into the constructed search engine to update the data stored in the constructed search engine.

3. The method according to claim 2, wherein, Importing data from the original data table into the constructed search engine includes: Create a temporary data table containing the fields from the constructed search engine; Import the product data from the original data table into the temporary data table; and Import the product data from the temporary data table into the constructed search engine in batches.

4. The method according to claim 2, wherein, The search engine is constructed through the following steps: The parameters of the distributed search engine are configured based on the magnitude of the product data from the e-commerce platform; and A distributed search engine is constructed based on the configured parameters, with the number of cluster nodes matching the magnitude of the product data.

5. The method according to claim 1, wherein, Based on the data stored in the constructed search engine, retrieve product information that matches the name of the target reverse product, and annotate the attributes of the target reverse product based on the matching product information, including: Based on the name of the target reverse product, the data stored in the constructed search engine is matched to recall product information that matches the name of the target reverse product; The recalled product information is sorted in descending order of time; and The attributes of the target reverse product are labeled based on a preset number of product information that are ranked first, wherein the attributes of the target reverse product include the product category.

6. The method according to claim 1, wherein, Quantifying the attributes of the target reverse product and adjusting the preset price of the target reverse product based on the quantification results to price the target reverse product, including: The attributes of the target reverse product are quantified; Based on the quantification results, optimized parameters matching the sales method are determined; and The preset price of the target reverse product is corrected based on the optimization parameters, and the target reverse product is priced at the corrected price.

7. The method according to claim 6, wherein, Based on the quantification results, optimized parameters matching the sales method are determined, including: Based on the quantification results, determine the auction success probability and auction failure probability that match the auction method; Based on the optimized parameters, the preset price of the target reverse product is adjusted, and the target reverse product is priced at the adjusted price, including: The starting price of the target reverse commodity is obtained by adjusting the minimum starting price of the target reverse commodity based on the auction success probability and the auction failure probability.

8. A search device based on reverse scenario, comprising: The attribute labeling module is configured to retrieve product information that matches the name of the target reverse product based on the data stored in the constructed search engine, and to label the attributes of the target reverse product based on the matching product information. as well as The product pricing module is configured to quantify the attributes of the target reverse product and, based on the quantification results, adjust the preset price of the target reverse product to price the target reverse product.

9. A search engine cluster comprising multiple nodes, each node comprising: One or more processors; Memory, used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors of the plurality of nodes implement the method as described in any one of claims 1-7.

10. A computer-readable medium having a computer program stored thereon that, when executed by a processor, implements the method as described in any one of claims 1-7.

Images