A jewelry dynamic pricing system and method

Abstract

The application discloses a kind of jewelry dynamic quotation system and method, it is related to jewelry transaction data processing technical field;The method comprises the following steps: collecting jewelry live transaction record and corresponding promotion benefit data, generating promotion context anchor;According to the division of the same attribute price group of commodity core attribute and the calculation of specification normalization transaction price, determine normal reference price;Combined with order state, promotion context anchor and price relationship, identify the low price transaction residual image after live promotion benefit detachment;For the transaction record that can determine promotion benefit amount, promotion restoration is carried out, form normal quotation sample set, and generate jewelry normal dynamic quotation accordingly.The application can avoid live short cycle promotion low price to lower normal quotation, improve quotation accuracy and credibility.

Description

Technical Field

[0001] This invention relates to the field of jewelry transaction data processing technology, and in particular to a dynamic pricing system and method for jewelry. Background Technology

[0002] With the development of online jewelry trading, the channels for jewelry transactions have gradually expanded from traditional wholesale, offline stores, and ordinary e-commerce platforms to real-time interactive trading scenarios such as live-streaming e-commerce and social e-commerce. During live-streaming sales, gold jewelry, karat gold jewelry, diamond jewelry, colored gemstone jewelry, and pearl jewelry are often combined with short-term promotional benefits such as live-streaming coupons, streamer subsidies, platform red envelopes, processing fee reductions, and gift discounts. This results in significant price differences for the same piece of jewelry across different live-streaming sessions, promotional periods, and combinations of benefits. Existing jewelry pricing systems primarily generate prices based on transaction prices, product attributes, and market popularity data, which can reflect jewelry trading trends to some extent. However, when live-streaming e-commerce transaction data is included in the pricing sample, it easily mixes short-term promotional transaction prices with normal transaction prices, leading to pricing results that fail to accurately reflect the stable market price of jewelry under non-promotional conditions.

[0003] Especially after the live stream ends, promotional links, coupon rules, host subsidy instructions, gift discount information, or processing fee reduction rules may become invalid, removed from the platform, or untraceable. However, the actual payment price in the transaction record remains as the real transaction price in the transaction data. These transaction records have the status of real payment, real fulfillment, or real order, but the reason for their low price has become detached from the context of promotional rights, forming a lingering low-price transaction after the live stream promotional rights have been detached. Existing pricing methods usually focus on identifying fake transactions, abnormal transactions, or price deviation data, making it difficult to distinguish between genuine normal transactions, reproducible promotional transactions, and genuine low-price transaction records where the promotional reason is untraceable. It is easy to directly use the latter as a sample of normal pricing, which in turn causes the dynamic price of jewelry to be continuously lowered by the short-term promotional price of the live stream, reducing the credibility of the pricing results and the reference value of the transaction. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies in live-stream e-commerce, where the transaction price of jewelry is easily separated from the context of promotional benefits, leading to the misuse of genuine low-price transaction records as normal price samples and reducing the accuracy of dynamic pricing. Therefore, this invention proposes a dynamic pricing system and method for jewelry.

[0005] To address the problems existing in the prior art, the present invention adopts the following technical solution: A method for dynamic pricing of jewelry includes: Step 1: Collect jewelry livestream sales records and their corresponding promotional benefits data, and bind the promotional benefits data with the jewelry livestream sales records to obtain the promotional context anchor; Step 2: Divide the products into price groups with the same attributes based on the core attributes of the jewelry live-streaming transaction records, and obtain the specification-normalized transaction price for each transaction record; Step 3: Determine the normal reference price within the same attribute price group; Step 4: Generate a low-price transaction afterimage marker based on the order status, the normalized transaction price, the normal reference price, and the promotional context anchor; Step 5: Filter transaction records based on the low-price transaction afterimage markers, and restore the promotion of transaction records with promotion context anchors that have a definite amount of promotional benefits to form a normal price sample set; Step 6: Generate a dynamic price quote for jewelry based on the set of normal price quote samples.

[0006] Preferably, the jewelry live-stream transaction record includes a product identifier, core product attributes, live-stream session identifier, transaction time, actual payment price, and order status; the core product attributes include at least one of the following for determining the price group of the same attribute: material, purity, weight, main stone type, main stone weight, certificate grade, and craftsmanship fee identifier; wherein, jewelry priced by weight includes at least weight, and jewelry containing a main stone includes at least the main stone type and main stone weight.

[0007] Preferably, the promotional context anchor includes a live broadcast session identifier, a valid promotional period, a promotional benefit amount, or a promotional benefit calculation rule; the promotional context anchor is bound to the jewelry live broadcast transaction record through the live broadcast session identifier and the transaction time.

[0008] Preferably, the price groups are divided according to the core attributes of the products in the live-streamed jewelry sales records, including: classifying sales records with the same material and purity into the same candidate price group; for sales records with certificate grades, further dividing them according to certificate grades within the candidate price group; for sales records with labor cost markings, further dividing them according to labor cost markings within the candidate price group; for jewelry priced by weight, further dividing them according to weight range within the candidate price group; and for jewelry containing a main stone, further dividing them according to the type and weight range of the main stone within the candidate price group.

[0009] Preferably, the specification-normalized transaction price for each transaction record is obtained as follows: for jewelry priced by weight, the ratio of the actual payment price to the actual weight is used as the specification-normalized transaction price; for jewelry priced by piece, the actual payment price within the same attribute price group is used as the specification-normalized transaction price.

[0010] Preferably, determining a normal reference price within the same attribute price group includes: selecting transaction records in the same attribute price group whose order status indicates the transaction record is a real transaction and which are not bound to a promotional context anchor, or transaction records in the same attribute price group whose order status indicates the transaction record is a real transaction and which are bound to a promotional context anchor but whose transaction time is not within the effective promotional period recorded by the promotional context anchor, as normal reference samples; and determining the median of the normal reference sample's specification-normalized transaction prices as the normal reference price of the same attribute price group.

[0011] Preferably, a low-price transaction afterimage marker is generated based on the order status, the specification-normalized transaction price, the normal reference price, and the promotion context anchor. This includes: when the order status indicates that the transaction record is a real transaction, the specification-normalized transaction price of the transaction record is lower than the normal reference price of its corresponding price group, and the promotion context anchor bound to the transaction record does not contain the promotion validity period, or does not contain the promotion benefit amount and the promotion benefit amount cannot be determined by the promotion benefit calculation rules, the transaction record is marked as a low-price transaction afterimage record.

[0012] Preferably, the process of filtering transaction records based on the low-price transaction afterimage markers and forming a normal price quotation sample set includes: excluding transaction records marked as low-price transaction afterimage records from the normal price quotation sample set; adding the normal price quotation sample set to ...

[0013] Preferably, the promotion restoration includes: for transaction records with a definite promotional benefit amount and not marked as low-price transaction afterimage records, adding the actual payment price to the promotional benefit amount to obtain the promotional restoration price; for jewelry priced by weight, adding the ratio of the promotional restoration price to the actual weight to the normal price sample set; for jewelry priced by piece, adding the promotional restoration price to the normal price sample set; generating a normal dynamic price for jewelry based on the normal price sample set, including: determining the median of the price samples in the normal price sample set as the normal dynamic price for jewelry in the same attribute price group, and outputting the normal dynamic price for jewelry and the low-price transaction afterimage mark.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention generates promotional context anchors by binding jewelry livestream transaction records with corresponding promotional benefit data. This allows short-term promotional benefits such as livestream coupons, streamer subsidies, platform red envelopes, processing fee reductions, and gift discounts to establish a traceable relationship with specific transaction records. Furthermore, it divides price groups based on the core attributes of the product, calculates standardized transaction prices and normal reference prices according to specifications, and prevents jewelry of different materials, purity, weight, main stone type, main stone weight, certificate grade, or processing fee markings from being directly mixed in price. In the multi-channel, high-promotion livestream e-commerce transaction environment, it can perform attribute consistency processing and promotional context restoration for jewelry transaction prices, avoiding the mixing of short-term promotional prices during livestreams with normal transaction prices, and improving the ability of dynamic jewelry pricing to reflect stable circulation prices.

[0015] 2. This invention identifies low-price transaction remnants after the detachment of live-stream promotional benefits by jointly identifying order status, standardized transaction price, normal reference price, and promotional context anchor. For genuine transaction records where the amount of promotional benefits can be determined, the promotion is restored and included in the normal price sample set. For genuine transactions where the effective time of the promotion or the amount of the promotional benefits cannot be traced back and is lower than the normal reference price, the transaction records are marked as low-price transaction remnants and excluded from the normal price sample set. While preserving the reproducible value of promotional transactions, this invention prevents live-stream low prices with untraceable promotional reasons from continuously lowering the normal dynamic price of jewelry, thereby improving the accuracy, credibility, and transaction reference value of the pricing results. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating a dynamic pricing method for jewelry provided in an embodiment of the present invention. Detailed Implementation

[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0018] Example: This invention provides a dynamic pricing system and method for jewelry, applicable to gold jewelry, platinum jewelry, K-gold jewelry, diamond jewelry, colored gemstone jewelry, pearl jewelry, jade jewelry, and other jewelry products traded through live-streaming e-commerce channels. The dynamic pricing system collects live-streaming transaction records and corresponding promotional rights data from live-streaming e-commerce channels, binds the promotional rights data with the live-streaming transaction records to form a promotional context anchor, and identifies low-price transaction remnants after the promotional rights are detached based on the promotional context anchor. This prevents genuine low-price orders from being untraceable due to promotional reasons from entering the normal pricing calculation, thus avoiding the normal dynamic pricing of jewelry being continuously lowered by short-term promotional prices during live-streaming. In this embodiment, the "low-price transaction residue" after the live-stream promotional benefits are removed refers to the actual payment price in the jewelry live-stream transaction record, which is formed by the superposition of at least one of the promotional benefits such as live-stream limited-time coupons, anchor subsidies, platform red envelopes, labor fee reductions, and gift discounts. However, after the live stream ends, the promotional link is removed, the promotional rules become invalid, or the promotional benefit information cannot be traced back, the actual payment price is still retained in the transaction record, and the data of the promotional benefits that formed the low price cannot be restored or verified by the system. Such transaction records may have a real order status, a real payment status, and a real fulfillment status. The problem is not whether the transaction is genuine or not, but that the actual payment price is not suitable as a quotation sample for the normal market circulation price.

[0019] Example 1: As Figure 1 As shown in the figure, this embodiment provides a dynamic pricing method for jewelry, which is executed by a dynamic pricing system for jewelry. The dynamic pricing system for jewelry includes a data acquisition terminal, a data processing terminal, and a pricing output terminal. The data acquisition terminal is used to collect live-streamed sales records and promotional benefit data for jewelry. The data processing terminal is used to generate promotional context anchors, divide price groups with the same attributes, calculate the normalized transaction price for specifications, determine the normal reference price, identify low-price transaction afterimage markers, and form a normal pricing sample set. The pricing output terminal is used to output the normal dynamic pricing of jewelry and the low-price transaction afterimage markers. The method described in this embodiment includes the following steps; S1. Collect jewelry live-stream transaction records and their corresponding promotional benefits data, and bind the promotional benefits data with the jewelry live-stream transaction records to obtain the promotional context anchor; When a transaction occurs during a live stream, the system collects and records the jewelry transaction history. ,in Indicates the first Transaction records; the aforementioned jewelry livestream transaction records It should include at least the product identifier, core product attributes, live stream session identifier, transaction time, actual payment price, and order status; Product identification is used to distinguish different jewelry items. Product identification can be a product code, platform product number, merchant item number, or a unique product identifier generated internally by the system. Product core attributes describe the attributes that affect the formation of jewelry prices. Product core attributes include at least one of the following to determine the price group of the same attribute: material, purity, carat weight, main stone type, main stone weight, certificate grade, and craftsmanship fee. Among them, jewelry priced by weight must include at least carat weight, and jewelry containing a main stone must include at least the main stone type and main stone weight. Live broadcast session identifier is used to identify the live broadcast session to which the transaction record belongs. Transaction time is used to identify the actual time when the transaction record was generated. Actual payment price is used to represent the order price actually paid by the consumer after deducting discounts, credits, or subsidies. Order status is used to indicate whether the transaction record has reached a real transaction status. Order status includes at least one of the following: paid, shipped, signed for, completed, canceled, and refunded. In this embodiment, the order status indicating that the transaction record is a real transaction means that the order status of the transaction record is at least one of paid, shipped, signed for, or completed, and the transaction record is not in the cancelled, fully refunded, or returned status. For transaction records that have been cancelled, fully refunded, or returned, the system does not treat them as real transaction records and does not use them for normal reference price calculation, low-price transaction afterimage marking, and normal quotation sample set construction. For transactions with partial refunds, the system processes them based on the reason for the refund. When a partial refund corresponds to a returned item, a change in item specifications, or a price resettlement, the system does not add the transaction to the normal reference sample or the normal price sample set. When a partial refund does not change the transaction price of jewelry and the order status still indicates that the item has been fulfilled, the system uses the actual payment price after the partial refund as the actual payment price for the transaction. In this way, the system can avoid interference from order cancellations, refunds, or resettlement orders on the normal dynamic pricing.

[0020] Collecting live-stream sales records of jewelry. At the same time, the system collects the promotional benefits data corresponding to the transaction record; the promotional benefits data includes at least one of the following: live broadcast room coupons, anchor subsidies, platform red envelopes, service fee reductions, and gift discounts, and the corresponding benefit amount or benefit calculation rules; the promotional benefits data also includes the effective promotion time, which includes the promotion start time and the promotion end time; In this embodiment, the amount of promotional benefits is determined according to the order settlement details or the promotional benefits calculation rules; for live broadcast coupons, anchor subsidies, platform red envelopes, and labor fee reductions, the system determines the corresponding deduction amount in the order settlement details as the amount of promotional benefits; for discount promotions, the system determines the amount reduced in the order settlement details after the order meets the discount conditions as the amount of promotional benefits; for discount promotions, the system determines the amount of promotional benefits based on the difference between the settlement base price of the product before the discount and the actual settlement price after the discount. Regarding the discount of gifts, if the order settlement details record the discount amount of gifts, the system will determine the discount amount of gifts as the promotional benefit amount; if the order settlement details do not record the discount amount of gifts, and the discount amount of gifts cannot be calculated according to the promotional benefit calculation rules, the system will not include the gifts in the promotional benefit amount, and will treat the transaction record as a transaction record whose promotional benefit amount cannot be determined by the promotional benefit calculation rules.

[0021] For the same jewelry livestream transaction record containing multiple promotional benefit amounts, the system will sum the individual promotional benefit amounts to obtain the total promotional benefit amount for that transaction record. in, Indicates the first The amount of promotional benefits corresponding to each jewelry livestream transaction record. Indicates the first The first item in the live-stream sales record of jewelry. The amount corresponding to each promotional benefit This indicates the number of promotional benefits with a definite amount in the transaction record; In this embodiment, the effective promotion time is determined by the promotion start time and promotion end time in the promotion rights data; for cases where there are multiple promotion time periods in the same live broadcast session, the system reads the promotion start time and promotion end time of each promotion time period and matches the transaction time of the jewelry live broadcast transaction record with each promotion time period; when the transaction time falls within one of the promotion time periods, the system determines that promotion time period as the effective promotion time corresponding to the jewelry live broadcast transaction record. When the promotional benefits data only records the live broadcast session identifier but not the start and end times of the promotion, the system determines that the promotion context anchor does not include the effective promotion time. If the transaction time of a jewelry live broadcast transaction record does not fall within any promotional time period, the system will not recognize the transaction record as a transaction record within the effective promotion time. Thus, the system can clearly define the correspondence between transaction records and the effective promotion time when there are multiple rounds of limited-time flash sales, phased coupons, or time-segmented fee reductions in the live broadcast room.

[0022] The system will combine promotional benefits data with jewelry livestream sales records. Binding, obtaining the promotional context anchor, denoted as Promotional context anchor At a minimum, it should include the live stream session identifier, the effective promotion period, the amount of promotional benefits, or the rules for calculating promotional benefits; and promotional context anchors. By identifying the live stream session and transaction time, and comparing the live stream transaction records for jewelry. Corresponding binding; specifically, when jewelry livestream sales records... If the live stream session identifier matches the live stream session identifier in the promotional benefits data, and the transaction time falls within the valid promotional period recorded in the promotional benefits data, the system will match the promotional benefits data with the jewelry live stream transaction record. Binding; In one specific implementation, a live-streaming session sells pure gold bracelets; the live-streaming session is identified as follows: The promotion starts at [time]. The promotion ends at [time]. ;Order The transaction time is ,and The system will then link the promotional benefits data from the live stream session with the order details. Binding forms promotional context anchors When the promotional benefits data directly records the platform's red envelope amount, the streamer's subsidy amount, and the labor fee reduction amount, the promotional context anchor... Record the amount of promotional benefits; when the promotional benefits data includes rules for spending a certain amount to reduce the price, discount rules, or gift conversion rules, the promotional context anchor... Record the rules for calculating promotional benefits; Through this step, the system simultaneously saves the promotional rights conditions that form the actual payment price when the transaction record is generated, so that the transaction price and the reasons for the promotional rights are traceably linked, avoiding the loss of promotional rights data while only retaining the actual payment price after the live broadcast ends.

[0023] S2. Based on the core attributes of the products in the jewelry live-streaming sales records, divide the prices into price groups with the same attributes and obtain the normalized sales price for each sales record. The system reads the live-stream sales records of jewelry. The product's core attributes are used to categorize prices into price groups based on these core attributes, denoted as follows: Price groups with the same attributes It is a set of transaction records used for price comparison and normal dynamic quotation calculation, the price group with the same attribute The transaction records within the same period have the same or comparable core product attributes; In dividing price groups with the same attributes When processing transactions, the system first groups transaction records with the same material and purity into the same candidate price group; for transaction records with certificate grades, the system further subdivides them within the candidate price group according to the certificate grade; for transaction records with labor cost indicators, the system further subdivides them within the candidate price group according to the labor cost indicator; for jewelry priced by weight, the system further subdivides them within the candidate price group according to the weight range; for jewelry containing a main stone, the system further subdivides them within the candidate price group according to the type and weight range of the main stone. Among them, the weight range is preferentially adopted from the weight range already recorded in the product listing specification field; when the product listing specification field does not record the weight range but records the actual weight, the system uses the weight range corresponding to the same specification identifier in the merchant's inventory specification as the weight range; when neither the product listing specification field nor the merchant's inventory specification records the weight range, the system will classify the transaction records with the same actual weight into the same weight range. The weight range of the main stone is preferentially adopted from the weight range of the main stone already recorded in the product certificate information or product listing specification field; when the product certificate information or product listing specification field does not record the weight range of the main stone but records the actual weight of the main stone, the system uses the weight range of the main stone corresponding to the same specification identifier in the merchant's inventory specification as the weight range of the main stone; when the product certificate information, product listing specification field and merchant inventory specification do not record the weight range of the main stone, the system will classify the transaction records with the same actual main stone weight into the same weight range of the main stone. Using the above method, the system prioritizes using existing product specification data to determine price groups with the same attributes, ensuring that jewelry items used for price comparison within the same attribute price group have comparable weight or main stone specifications.

[0024] After grouping prices by attribute, the system calculates the standardized transaction price for each transaction record; for jewelry priced by weight, the system uses the ratio of the actual payment price to the actual weight as the standardized transaction price. in, Indicates the first The specifications of each transaction record are standardized based on the transaction price. Indicates the first The actual payment price of each transaction record. Indicates the first The actual weight of the jewelry corresponding to each transaction record; For jewelry priced per piece, the system uses the actual payment price within the same attribute price group as the standardized transaction price. In other words, when jewelry is not suitable for conversion by weight, such as a diamond ring of a fixed size, a pearl necklace of a fixed style, or a colored gemstone pendant of a fixed size, the system will use the actual payment price within the same attribute price group for subsequent comparison.

[0025] For jewelry priced per piece, the system only uses the actual payment price within the same attribute price group as the standardized transaction price after the transaction record has been classified into the same attribute price group according to the corresponding core attributes of the product in terms of material, purity, main stone type, main stone weight range, certificate grade, and craftsmanship fee. In other words, jewelry priced per piece does not directly mix and compare all items priced per piece, but rather uses the actual payment price as the basis for determining the subsequent normal reference price and identifying low-price transaction shadows, provided that the core attributes of the products are already comparable. This step helps the system avoid directly mixing and comparing the transaction prices of jewelry with different materials, colors, weights, main stone types, main stone weights, or certificate grades, thereby improving the accuracy of subsequent low-price transaction afterimage identification.

[0026] S3. Determine the normal reference price within the same attribute price group; The system sets prices for each group with the same attribute. The standard reference price is determined in China and denoted as . Normal reference price Used to represent price groups with the same attribute Price reference before the separation of live-streaming promotional rights; In determining the normal reference price At that time, the system in the same attribute price group The following transaction records are selected as normal reference samples: those whose order status indicates that the transaction record is a real transaction and is not bound to a promotion context anchor, or those whose order status indicates that the transaction record is a real transaction and is bound to a promotion context anchor but whose transaction time is not within the effective time of the promotion recorded by the promotion context anchor. An order status indicating a genuine transaction means that the order status meets at least one of the following: paid, shipped, signed for, or completed, and is not in the cancelled, fully refunded, or returned status. For transaction records with partial refunds, their eligibility as genuine transactions is determined according to the partial refund processing rules in S1. For transaction records with partial refunds that do not change the transaction price of the jewelry and fulfill the contract, the system uses the actual payment price after the partial refund as the actual payment price and allows it to be used as a normal reference sample if other conditions in this step are met. For transaction records with partial refunds and corresponding product returns, product specification changes, or price resettlement, the system does not use them as normal reference samples. For transaction records that have a promotional context anchor but whose transaction time is not within the effective promotion period, the system will treat them as transaction records that are not affected by the promotional benefits recorded in the promotional context anchor and allow them to be included in the normal reference sample.

[0027] When there is no normal reference sample in the price group with the same attribute, the system does not generate a normal reference price for that price group. The system will temporarily refrain from marking low-price transaction shadows on transaction records within the same price group; the normal reference price will be determined once new normal reference samples that meet the conditions are added to the same price group. and based on the normal reference price Execute low-price transaction afterimage marking; When a normal reference sample exists in a price group with the same attribute, the system determines the median of the normalized transaction price in the normal reference sample as the normal reference price for the price group with the same attribute. As a result, the system can avoid forcibly generating a normal reference price when there is no normal reference sample, and ensure that the low-price transaction afterimage mark has a comparable price benchmark.

[0028] The system determines the median of the normalized transaction prices in the normal reference sample as the normal reference price for the same attribute price group: in, Indicates price groups with the same attributes The normal reference price Indicates price groups with the same attributes The normalized transaction price based on the specifications of the standard reference sample in China; The median is a commonly used robust position statistic in statistics. Compared with the arithmetic mean, the median is less affected by a small number of extremely high or low prices, making it suitable for scenarios where there are promotional disturbances, short-term discounts, or a small number of abnormal transaction prices in the transaction price sample. In this embodiment, the median is used to determine the normal reference price, which can reduce the impact of short-term promotional prices during live streaming on the normal reference price. In one specific implementation, the price group with the same attribute is a pure gold bracelet price group, with the material being pure gold, the purity being pure gold, the weight range being the same specification range, and the processing fee identifier being the same; the system selects transaction records in this price group that are not bound to a promotional context anchor and whose order status is completed as normal reference samples, and calculates the median of their specification-normalized transaction prices to obtain the normal reference price for this pure gold bracelet price group. .

[0029] S4. Generate a low-price transaction afterimage marker based on order status, specifications, normalized transaction price, normal reference price, and promotional context anchor; The system normalizes the transaction price based on order status and specifications. Regular reference price and promotional context anchors To determine the transaction record of each jewelry live stream Whether it belongs to the low-price transaction after the promotional rights have been removed, and generate a low-price transaction afterimage marker, denoted as: ; For the Jewelry and accessories live streaming sales records The system marks a low-price transaction ghost record when it meets the following conditions simultaneously: The order status indicates that the transaction record is a genuine transaction; specifically, the order status is at least one of paid, shipped, signed for, or completed, and is not cancelled, fully refunded, or returned; for transaction records with partial refunds, whether they belong to genuine transaction records and the actual payment price are determined in accordance with the partial refund processing rules in S1. The normalized transaction price of this transaction record It is lower than the normal reference price of its corresponding price group G. This condition indicates that the transaction record exhibits a price performance below the normal reference price within the same price group. In this embodiment, the standardized transaction price is... Below the normal reference price This is only used to identify the transaction record as a candidate low-price transaction record, and is not used as the sole basis for determining low-price transaction ghost records; the system also needs to consider the order status and promotional context. Further judgment is needed to determine whether the amount of promotional benefits can be determined; only if the transaction record is a genuine transaction and the promotional context anchor... The system will only mark a transaction record as a low-price transaction ghost record if it does not include the effective period of the promotion, or does not include the amount of promotional benefits and the amount of promotional benefits cannot be determined by the promotional benefits calculation rules. The promotional context anchor attached to this transaction record The transaction does not include the promotion's validity period, or it does not include the amount of promotional benefits and cannot be determined using the promotional benefits calculation rules; this condition indicates that although the transaction record shows a low price, the system cannot determine the amount of promotional benefits through the promotional context anchor. Explain the reasons behind the promotional benefits that resulted in this low price; When all three conditions above are met, the system generates a low-price transaction afterimage marker. , indicating the first This transaction record represents a low-price transaction remnant after promotional benefits have been removed; when none of the above three conditions are met simultaneously, the system generates a low-price transaction remnant marker. , indicating the first The transaction record is not a remnant of a low-price transaction after the promotional rights were removed.

[0030] It should be noted that the system does not automatically recognize all low-price transaction records as low-price transaction shadow records; if a low-price transaction record is lower than the normal reference price, but its promotional context anchor... If a transaction record includes the effective promotion period and the amount of promotional benefits, or includes the effective promotion period and the rules for calculating the amount of promotional benefits, then the transaction record can be restored by the system and will not be marked as a low-price transaction remnant record; if a transaction record is not a real transaction, such as an order that has been cancelled, fully refunded, or returned, then the transaction record will not be marked as a real low-price transaction remnant in this step, but will be filtered out by the order status. Through this step, the system can distinguish three types of transaction data: the first type is normal transaction records; the second type is live transaction records with reproducible promotional benefits; and the third type is low-price transaction remnant records where the actual transaction was completed but the promotional benefits cannot be reproduced. This distinction can prevent the direct use of real low-price orders with untraceable promotional reasons in normal price calculations.

[0031] S5. Filter transaction records based on the low-price transaction afterimage marker, and restore the promotion of transaction records with promotion context anchors that have a definite amount of promotional benefits to form a normal price sample set. The system uses low-price transaction afterimage markings The transaction records are filtered to form a sample set of normal price quotes, denoted as . Standard price quotation sample set It is a price sample set used to calculate the normal dynamic price of jewelry in the same attribute price group G; The system generates a set of normal price quotation samples according to the following rules. ; when At that time, the system will exclude this transaction record from the normal quotation sample set. In addition; although the transaction record is a real transaction, its actual payment price is lower than the normal reference price, and the promotional context anchor cannot restore the amount of promotional benefits that formed the low price. Therefore, its actual payment price is not suitable for calculating the normal dynamic price. when Furthermore, if the transaction record is not bound to a promotional context anchor, the system will normalize the transaction record's specifications to a transaction price. Add to the normal price sample set ; when Furthermore, when the transaction record has a promotional context anchor that can determine the amount of promotional benefits, the system will restore the promotion for the transaction record. A promotional context anchor that can determine the amount of promotional benefits means that the promotional context anchor includes the effective time of the promotion and the amount of promotional benefits, or includes the effective time of the promotion and the promotional benefit calculation rules that can calculate the amount of promotional benefits.

[0032] Promotional restoration refers to the system obtaining a promotional restoration price based on the actual payment price and the amount of promotional benefits. In this embodiment, the promotional restoration price represents the price base of the transaction record before deducting the live-stream promotional benefits. The system only performs promotional restoration on transaction records where the amount of promotional benefits can be determined. For transaction records where the amount of promotional benefits cannot be determined, the system does not perform promotional restoration. If the same transaction record has multiple promotional benefits with definite amounts, the system sums the amounts of each promotional benefit and adds them to the actual payment price to obtain the promotional restoration price. The promotional restoration price is not used to change the original order amount, but only to construct a normal price sample set. Specifically, the system adds the actual payment price to the promotional benefit amount to obtain the promotional restoration price: in, Indicates the first Promotional price restoration for each transaction record. Indicates the first The actual payment price of each transaction record. Indicates anchor based on promotional context. The amount of promotional benefits that can be determined.

[0033] For jewelry priced by weight, the system adds the ratio of the promotional price to the actual weight to the regular price sample set. : in, This indicates the price after the promotional period has been restored to a uniform price for the same specifications. This indicates the promotional price has been restored. Indicates actual weight in grams; For jewelry priced per piece, the system will restore the promotional price. Add to the normal price sample set ; In one specific implementation, the actual payment price of a certain pure gold bracelet is Promotional context anchor The amount of promotional benefits recorded in the document is The actual weight is The system first calculates the promotional price. Then calculate the unified price for the specifications after the promotion. and will Add to the normal price sample set If the actual payment price of another pure gold bracelet transaction is significantly lower than the normal reference price, but the promotional context anchor lacks the amount of promotional benefits, and the amount of promotional benefits cannot be determined through the promotional benefits calculation rules, then this transaction record will be marked as a low-price transaction ghost record and will not be included in the normal price sample set. ; This step involves not simply deleting all live-stream promotional orders, but rather restoring the promotional benefits to transaction records where the promotional benefit amount can be recovered, and using the restored price as a sample for normal pricing. Transaction records with genuinely low prices that cannot be restored to their promotional benefit amount and are lower than the normal reference price are excluded, thus reducing the impact of the lingering effects of low-price transactions after the live-stream promotional benefits are removed on normal dynamic pricing.

[0034] S6. Generate dynamic price quotes for jewelry based on a set of normal price quote samples; The system is based on a normal price quotation sample set. Generate a normal dynamic price quote for jewelry items in the same price group, denoted as . ; When the normal price sample set When the value is empty, the system will not generate a normal dynamic price quote for jewelry in the same attribute price group G. And retain the generated low-price transaction afterimage markers. Once the same price group G has any new transaction records that are not marked as low-price transaction remnants, or new transaction records that can be restored through promotional context anchors, the system will then form a non-empty normal price sample set. And generate regular dynamic prices for jewelry. ; When the normal price sample set When not empty, the system will use the normal quotation sample set. The median of the price sample was determined as the normal dynamic price of jewelry in the same price group.

[0035] The system will use a set of normal price quotes. The median of the price sample was determined as the normal dynamic price of jewelry in the same price group: in, Indicates price groups with the same attributes Regular dynamic pricing for jewelry. Represents the normal price quotation sample set The k-th price sample in the data; The system outputs regular dynamic price quotes for jewelry. and low-price transaction afterimage mark Among them, the regular dynamic price of jewelry. Used to indicate the stable market price after removing the impact of promotional benefits and low-price afterimages; low-price transaction afterimage marker. Used to identify genuine low-price transaction records that are not included in the normal price calculation; In one specific implementation, price groups with the same attribute The system sets prices for pure gold bracelets of the same material, purity, weight range, and processing fee; it also adds price samples that are not bound to promotional context anchors and whose order status is a real transaction, as well as price samples whose promotional benefits can be restored and which are not marked as low-price transaction afterimages, to the normal price sample set. Then calculate the normal price sample set. The median of the price sample is used to obtain the normal dynamic price of jewelry in this pure gold bracelet price group. .

[0036] Example 2: This example provides a dynamic pricing system for jewelry, which is used to execute the dynamic pricing method for jewelry described in Example 1; the system includes a transaction record collection module, a promotional context anchor generation module, a specification normalization module, a normal reference price calculation module, a low-price transaction afterimage recognition module, a normal price sample construction module, and a dynamic price generation module; The transaction record collection module is used to collect transaction records from live jewelry sales. The jewelry livestream sales records This includes product identification, core product attributes, live stream session identification, transaction time, actual payment price, and order status; core product attributes include at least one of the following used to determine the price group of the same attribute: material, purity, weight, main stone type, main stone weight, certificate grade, and labor cost identification; the transaction record collection module is also used to determine whether the transaction record is a real transaction record based on the order status, and to determine whether the actual payment price can be used for subsequent calculations according to the order status and refund order processing rules in Example 1; The promotional context anchor generation module is used to collect promotional benefit data corresponding to jewelry livestream sales records, and bind the promotional benefit data with the jewelry livestream sales records to generate promotional context anchors. The promotional context anchor This includes the live stream session identifier, the effective promotion period, and the promotional benefit amount or promotional benefit calculation rules; the promotional context anchor generation module binds the promotional benefit data to the jewelry live stream transaction record through the live stream session identifier and the transaction time; the promotional context anchor generation module is also used to determine the promotional benefit amount based on the order settlement details or promotional benefit calculation rules, and when there are multiple promotional benefits with definite amounts in the same transaction record, the sum of the amounts of each promotional benefit is used to obtain the promotional benefit amount corresponding to the transaction record; The specification normalization module is used to divide products into price groups with the same attributes based on the core attributes of the products in the live-streaming sales records of jewelry. And obtain the standardized transaction price for each transaction record. The specification normalization module first groups transaction records with the same material and purity into the same candidate price group. For transaction records with certificate grades, the candidate price group is further subdivided according to the certificate grade. For transaction records with labor cost indicators, the candidate price group is further subdivided according to the labor cost indicator. For jewelry priced by weight, the candidate price group is further subdivided into price groups of the same attribute according to the weight range. For jewelry containing a main stone, the candidate price group is further subdivided into price groups of the same attribute according to the type and weight range of the main stone. For jewelry priced by weight, the specification normalization module uses the ratio of the actual payment price to the actual weight as the specification normalized transaction price. For jewelry priced by piece, the specification normalization module uses the actual payment price within the same attribute price group as the specification normalized transaction price. The standard reference price calculation module is used for price groups with the same attribute. China establishes a normal reference price The normal reference price calculation module selects transaction records from the same attribute price group whose order status indicates the transaction record is a real transaction and is not bound to a promotional context anchor, or whose order status indicates the transaction record is a real transaction and is bound to a promotional context anchor but the transaction time is not within the effective time of the promotion recorded by the promotional context anchor, as normal reference samples; then, the median of the specification-normalized transaction prices in the normal reference samples is determined as the normal reference price of the same attribute price group. When there is no normal reference sample in the price group with the same attribute, the normal reference price calculation module does not generate a normal reference price for that price group with the same attribute. And the low-price transaction afterimage recognition module will temporarily not perform low-price transaction afterimage marking on transaction records in the same price group; The low-price transaction afterimage recognition module is used to normalize the transaction price based on order status and specifications. Regular reference price and promotional context anchors Generate low-price transaction afterimage markers When the order status indicates that the transaction record is a genuine transaction, the specifications of that transaction record are normalized to the transaction price. Lower than the normal reference price of its corresponding price group And the promotional context anchor bound to this transaction record If the transaction does not include the promotion's validity period, or does not include the amount of promotional benefits and the amount of promotional benefits cannot be determined according to the promotional benefits calculation rules, the low-price transaction afterimage recognition module marks the transaction record as a low-price transaction afterimage record and generates a low-price transaction afterimage marker. Otherwise, the low-price transaction afterimage recognition module generates a low-price transaction afterimage marker. ; The normal price quotation sample construction module is used to mark low-price transaction afterimages. Filter transaction records and perform promotion restoration on transaction records with identifiable promotional benefit amounts to form a normal pricing sample set. When a transaction record is marked as a low-price transaction ghost record, the normal price sample construction module excludes the transaction record from the normal price sample set; when a transaction record is not bound to a promotion context anchor and is not marked as a low-price transaction ghost record, the normal price sample construction module adds the normal price sample to the normal price sample set; when a transaction record has a promotion context anchor with a definite promotion benefit amount and is not marked as a low-price transaction ghost record, the normal price sample construction module adds the actual payment price to the promotion benefit amount to obtain the promotion restored price, and adds the promotion restored price sample to the normal price sample set. The dynamic quote generation module is used to generate quotes based on a set of normal quote samples. Generate regular dynamic pricing for jewelry. When the normal price sample set When empty, the dynamic price generation module will not generate a normal dynamic price for jewelry in the same price group. And retain the generated low-price transaction afterimage markers. When the normal price sample set When not empty, the dynamic quote generation module will generate a normal quote sample set. The median of the price sample was determined as the normal dynamic price of jewelry in the same attribute price group G. It also provides regular and dynamic price quotes for jewelry. and low-price transaction afterimage mark ; The system described in this embodiment is deployed on a server, cloud computing platform, local computing device, or merchant quotation management terminal; the transaction record collection module, promotion context anchor generation module, specification normalization module, normal reference price calculation module, low-price transaction afterimage recognition module, normal quotation sample construction module, and dynamic quotation generation module are implemented by software programs, or by combining software programs with general-purpose computing hardware.

[0037] Example 3: This example uses a pure gold bracelet as an example to illustrate the specific application process of the present invention; a live broadcast session sells pure gold bracelets, and the core attributes of the product include material, purity, weight, and processing fee; the system collects a jewelry live broadcast transaction record. The product identifier in this record is The live broadcast session is marked as The transaction time is The actual price paid is The actual weight is The order status is now complete. The system also collects promotional benefits data for this live stream session; if the promotional benefits data records a valid promotional period of... to ,and The system will link this promotional benefits data with the jewelry livestream sales records. Binding, generating promotional context anchors If the promotional context anchor The amount includes platform red envelopes, streamer subsidies, and fee reductions. The system will then aggregate these amounts into a total promotional benefit amount. If the promotional context anchor This includes rules for discounts or minimum purchase amounts; the system calculates the promotional benefits amount based on these rules and the order settlement details. If the promotional context anchor If the transaction only records the information of the gift but not the value of the gift, and the value of the gift cannot be determined according to the rules for calculating promotional benefits, the system will not include the gift in the amount of promotional benefits and will treat the transaction as a transaction record for which the amount of promotional benefits cannot be determined according to the rules for calculating promotional benefits. The system records live-streamed jewelry sales based on material, purity, weight range, and processing fee. Classified into price groups with the same attributes System calculation Standardized transaction price based on specifications: The system in price groups with the same attributes Transaction records with order status indicating a genuine transaction and not bound to a promotional context anchor, or transaction records with order status indicating a genuine transaction and bound to a promotional context anchor but whose transaction time is not within the promotion's effective period, are selected as the normal reference sample; if the same attribute price group Since there are no normal reference samples, the system does not generate price groups with the same attribute. Normal reference price And temporarily not for this price group with the same attribute The transaction records in the data are marked with a low-price transaction afterimage; if the price groups of the same attribute are... There are normal reference samples. The system calculates the median of the normalized transaction prices of these normal reference samples to obtain the normal reference price. ; like Then the system continues to determine the promotion context anchor. Can the amount of promotional benefits be determined? Includes the promotion's validity period and the amount of promotional benefits. The system will not Instead of marking it as a low-price transaction record, the actual price paid will be recorded. With promotional benefits amount Add them together to get the promotional price: The system then calculates the normalized price for the specifications after the promotion is restored: And Add to same attribute price group Normal price sample set ; If another jewelry live stream sales record The order status is "completed," and the specifications have been standardized with a fixed transaction price. Below the normal reference price However, its promotional context anchor If the promotion does not include the effective promotion period, or does not include the promotional benefit amount and the promotional benefit amount cannot be determined according to the promotional benefit calculation rules, then the system will... Records marked as low-price transaction afterimages, i.e. This transaction record Although it is a real transaction, it is not included in the normal price sample set. ; If the normal quotation sample set If empty, the system will not generate a price group for the same attribute of a pure gold bracelet. Regular dynamic price of jewelry And retain the generated low-price transaction afterimage markers; if the normal price sample set Not empty, the system is based on the normal quotation sample set. The median was calculated from the price sample to obtain the price group of pure gold bracelets with the same attributes. Regular dynamic price of jewelry It also outputs a low-price transaction afterimage marker. and ;in, Representing records It's not a record of a low-price transaction. Representing records It is a record of the lingering effects of low-price transactions after promotional rights have been removed; Through the above process, the system retains the real transaction records of the live broadcast that can be restored by promotions, while excluding the real low-price transaction records that cannot be restored after the promotion rights are removed, thereby preventing the normal dynamic price of the pure gold bracelet from being continuously dragged down by the low price of the short-term live broadcast.

[0038] Example 4: This example uses a fixed-size diamond ring as an example to illustrate the application of this invention in jewelry priced per piece; a live-streaming session sells fixed-size diamond rings, whose core attributes include material, main stone type, main stone weight, certificate grade, and craftsmanship fee; the system collects the jewelry live-streaming sales records. The actual price paid was The order status is now complete. The system records live-streamed jewelry sales based on material, main stone type, main stone weight range, certificate grade, and craftsmanship fee. Classified into price groups with the same attributes Since this diamond ring is priced per piece, the system has already categorized it into the same price group based on its material, main stone type, main stone weight range, certificate grade, and craftsmanship fee. Then, group the prices of items with the same attribute. The actual payment price is used as the standardized transaction price, i.e. ; The system in price groups with the same attributes China establishes a normal reference price If the price group has the same attribute Since there are no normal reference samples, the system does not generate price groups with the same attribute. Normal reference price And temporarily not for this price group with the same attribute The transaction records in the data are marked with a low-price transaction afterimage; if the price groups of the same attribute are... The system uses a normalized reference sample as its reference price, and determines the median of the normalized transaction prices of the normalized reference sample as the normalized reference price. ; like Furthermore, the sales records of jewelry live streams Promotional context anchor If the amount of promotional benefits is not included, and the amount of promotional benefits cannot be determined through the promotional benefits calculation rules, then the system will... Records marked as low-price transaction afterimages are not included in the normal price quotation sample set. ; If another diamond ring's live-stream sales record Promotional context anchor Includes the promotion's validity period and the amount of promotional benefits. The system will then calculate the actual payment price. With promotional benefits amount Add them together to get the promotional price. and will Add to the normal price sample set ; When the normal price sample set When empty, the system does not generate a price group for diamond rings of the same attributes with fixed specifications. Regular dynamic price of jewelry When the normal price sample set When not empty, the system will use the normal quotation sample set. The median of the price sample was determined as the price group of diamond rings of the same attributes with fixed specifications. Regular dynamic price of jewelry Therefore, the system can adapt to pricing scenarios for inlaid jewelry that are priced per piece.

[0039] Example 5: In this example, the system creates a record table for each jewelry livestream transaction record; the record table includes a transaction record field, a promotion context anchor field, a price processing field, and a quotation result field; The transaction record fields include product identifier, product core attributes, live broadcast session identifier, transaction time, actual payment price, and order status; the promotion context anchor fields include live broadcast session identifier, promotion validity period, promotion benefit amount, and promotion benefit calculation rules; the price processing fields include same attribute price group identifier, specification-normalized transaction price, normal reference price, promotion restored price, and normal quotation sample set identifier; the quotation result fields include normal dynamic quotation for jewelry and low-price transaction afterimage marker; When the low-price transaction afterimage is marked as 1, the system records in the quotation result field that the transaction record has not entered the normal quotation sample set; when the low-price transaction afterimage is marked as 0 and the transaction record has a promotion context anchor that can determine the amount of promotional benefits, the system records the promotion restoration price in the price processing field and writes the promotion restoration price sample into the normal quotation sample set; when the low-price transaction afterimage is marked as 0 and the transaction record is not bound to a promotion context anchor, the system writes the specification-normalized transaction price into the normal quotation sample set. When the normal reference sample is empty, the system does not write the normal reference price into the price processing field, and records that the low-price transaction afterimage mark is not executed temporarily in the data status corresponding to the price group identifier of the same attribute; when the normal quotation sample set is empty, the system does not write the normal dynamic quotation of jewelry into the quotation result field, and retains the low-price transaction afterimage mark that has already been generated; after the normal reference sample or normal quotation sample is added later, the system will update the normal reference price, the low-price transaction afterimage mark and the normal dynamic quotation of jewelry. Through the above data storage and tagging output methods, the system can retain low-price transaction afterimage tags while generating normal dynamic price quotes for jewelry, which facilitates subsequent verification of the source of the price quote samples.

[0040] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for dynamic pricing of jewelry, characterized in that, Includes the following steps: Step 1: Collect jewelry livestream sales records and their corresponding promotional benefits data, and bind the promotional benefits data with the jewelry livestream sales records to obtain the promotional context anchor; Step 2: Divide the products into price groups with the same attributes based on the core attributes of the jewelry live-streaming transaction records, and obtain the specification-normalized transaction price for each transaction record; Step 3: Determine the normal reference price within the same attribute price group; Step 4: Generate a low-price transaction afterimage marker based on the order status, the normalized transaction price, the normal reference price, and the promotional context anchor; Step 5: Filter transaction records based on the low-price transaction afterimage markers, and restore the promotion of transaction records with promotion context anchors that have a definite amount of promotional benefits to form a normal price sample set; Step 6: Generate a dynamic price quote for jewelry based on the set of normal price quote samples.

2. The method for dynamic pricing of jewelry according to claim 1, characterized in that, The jewelry live-streaming transaction record includes product identification, core product attributes, live-streaming session identification, transaction time, actual payment price, and order status; the core product attributes include at least one of the following used to determine the price group of the same attribute: material, purity, weight, main stone type, main stone weight, certificate grade, and craftsmanship fee identification; wherein, jewelry priced by weight includes at least weight, and jewelry containing a main stone includes at least the main stone type and main stone weight.

3. The method for dynamic pricing of jewelry according to claim 2, characterized in that, The promotional context anchor includes a live broadcast session identifier, a valid promotion period, a promotional benefit amount, or a promotional benefit calculation rule; the promotional context anchor is bound to the jewelry live broadcast transaction record through the live broadcast session identifier and the transaction time.

4. The method for dynamic pricing of jewelry according to claim 2, characterized in that, Based on the core attributes of the jewelry sales records from the live stream, price groups with the same attributes are divided, including: sales records with the same material and purity are grouped into the same candidate price group; for sales records with certificate grades, the candidate price group is further divided according to the certificate grade; for sales records with craftsmanship fee markings, the candidate price group is further divided according to the craftsmanship fee markings; for jewelry priced by weight, the candidate price group is further divided according to the weight range; for jewelry containing a main stone, the candidate price group is further divided according to the type and weight range of the main stone.

5. The method for dynamic pricing of jewelry according to claim 2, characterized in that, Obtain the standardized transaction price for each transaction record, including: for jewelry priced by weight, the ratio of the actual payment price to the actual weight is used as the standardized transaction price; for jewelry priced by piece, the actual payment price within the same price group is used as the standardized transaction price.

6. The method for dynamic pricing of jewelry according to claim 3, characterized in that, Determining a normal reference price within the same attribute price group includes: selecting transaction records from the same attribute price group whose order status indicates the transaction record is a real transaction and which are not bound to a promotional context anchor, or whose order status indicates the transaction record is a real transaction and which are bound to a promotional context anchor but whose transaction time is not within the effective promotional period recorded by the promotional context anchor, as normal reference samples; and determining the median of the specification-normalized transaction prices in the normal reference samples as the normal reference price of the same attribute price group.

7. The method for dynamic pricing of jewelry according to claim 6, characterized in that, Based on the order status, the normalized transaction price, the normal reference price, and the promotion context anchor, a low-price transaction afterimage marker is generated, including: when the order status indicates that the transaction record is a real transaction, the normalized transaction price of the transaction record is lower than the normal reference price of its corresponding price group, and the promotion context anchor bound to the transaction record does not contain the promotion validity period, or does not contain the promotion benefit amount and the promotion benefit amount cannot be determined by the promotion benefit calculation rules, the transaction record is marked as a low-price transaction afterimage record.

8. The method for dynamic pricing of jewelry according to claim 7, characterized in that, Based on the low-price transaction afterimage markers, transaction records are filtered to form a normal price sample set, including: excluding transaction records marked as low-price transaction afterimage records from the normal price sample set; adding the normal price sample set to ...

9. The method for dynamic pricing of jewelry according to claim 8, characterized in that, The promotion restoration includes: for transaction records with a definite promotional benefit amount and not marked as low-price transaction afterimage records, adding the actual payment price to the promotional benefit amount to obtain the promotional restoration price; for jewelry priced by weight, adding the ratio of the promotional restoration price to the actual weight to the normal price sample set; for jewelry priced by piece, adding the promotional restoration price to the normal price sample set; generating a normal dynamic price for jewelry based on the normal price sample set, including: determining the median of the price samples in the normal price sample set as the normal dynamic price for jewelry in the same attribute price group, and outputting the normal dynamic price for jewelry and the low-price transaction afterimage mark.

10. A dynamic pricing system for jewelry, used to execute the dynamic pricing method for jewelry as described in any one of claims 1-9, characterized in that, include: The module includes a transaction record collection module, a promotional context anchor generation module, a specification normalization module, a normal reference price calculation module, a low-price transaction afterimage recognition module, a normal quotation sample construction module, and a dynamic quotation generation module. The transaction record collection module collects live-streamed transaction records for jewelry, which include the core attributes of the product, the transaction time, the actual payment price, and the order status. The promotional context anchor generation module collects the promotional rights data corresponding to the jewelry live-streaming transaction records, and binds the promotional rights data with the jewelry live-streaming transaction records to obtain the promotional context anchor. The specification normalization module divides the product into price groups with the same attribute based on the product's core attributes and obtains the specification normalized transaction price for each transaction record. The normal reference price calculation module determines the normal reference price in the same attribute price group; The low-price transaction afterimage recognition module generates a low-price transaction afterimage marker based on the order status, the specification-normalized transaction price, the normal reference price, and the promotion context anchor. The normal price quotation sample construction module filters transaction records based on the low-price transaction afterimage markers and performs promotion restoration on transaction records with promotion context anchors that have a definite amount of promotional benefits, forming a normal price quotation sample set. The dynamic price generation module generates normal dynamic prices for jewelry based on the normal price sample set.

Images