Virtual item transaction data processing method and device

Abstract

The invention discloses a virtual item transaction data processing method and device. Information texts related to historical virtual article transactions are collected, and historical valuation reference information and historical user related information of historical transaction users are extracted from the information texts through a large language model; when the transaction of the current virtual article is detected, determining current evaluation reference information related to the current virtual article from the historical evaluation reference information, and determining an estimated transaction price of the current virtual article according to the current evaluation reference information; determining current user related information of two current transaction parties of the current virtual article transaction from the historical user related information; and verifying the current virtual article transaction according to the estimated transaction price, the to-be-transacted price of the current virtual article transaction and the current user related information, and outputting an obtained verification result. According to the scheme, the problem that the transaction judgment result is affected due to the fact that the evaluation accuracy is not high in existing virtual article transaction verification can be solved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and apparatus for processing virtual item transaction data. Background Technology

[0002] With the development of the internet, virtual items have gradually become digital assets with real economic value, thus creating a huge demand for virtual item trading. However, the complexity and risks of virtual item trading have also increased accordingly, with various illegal trading activities frequently occurring. Against this backdrop, accurate virtual item valuation and efficient compliance verification have become core requirements for online operations.

[0003] Existing methods for valuing virtual items estimate their value based on historical transaction prices, then compare these estimated prices with actual transaction prices to determine if any transactions are abnormal. However, this method struggles to address valuation biases arising from the complexity of virtual item attributes, frequent market fluctuations, and diverse transaction scenarios. In particular, the lack of historical data significantly reduces valuation accuracy when dealing with new types of virtual items or those with high rarity leading to limited historical transaction volumes.

[0004] Therefore, a method is needed to address the problem of low valuation accuracy in existing virtual item transaction verification, which affects the transaction judgment result. Summary of the Invention

[0005] This invention provides a method and apparatus for processing virtual item transaction data to solve the problem of low valuation accuracy in existing virtual item transaction verification, which affects the transaction judgment result.

[0006] According to one aspect of the present invention, a method for processing virtual item transaction data is provided, comprising:

[0007] Collect information text related to historical virtual item transactions, and extract historical valuation reference information and historical user-related information of historical transaction users from the information text using the large language model;

[0008] When a current virtual item transaction is detected, the current valuation reference information related to the current virtual item is determined from the historical valuation reference information, and the estimated transaction price of the current virtual item is determined based on the current valuation reference information; the current user information of the current trading parties of the current virtual item transaction is determined from the historical user information information.

[0009] The current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item transaction, and the current user's relevant information, and the verification result is output.

[0010] Optionally, the information text includes virtual commodity price discussion text, virtual item attribute change text, and virtual commodity price information text obtained from at least one message acquisition source; the historical valuation reference information includes price discussion information, price change trend information, and publicly available price information; correspondingly, extracting historical valuation reference information from the information text through the large language model includes:

[0011] The large language model is used to perform semantic analysis on the discussion text to extract first text content fragments containing virtual item price discussion information. The price discussion information of the virtual item is determined based on the text content about the price of the virtual item in each of the first text content fragments.

[0012] The large language model is used to perform semantic analysis on the text of the virtual item attribute change, and a second text content segment containing the attribute change value of the virtual item is extracted. The price change trend information of the virtual item is determined based on the virtual item attribute change value in each second text content segment.

[0013] The large language model is used to perform semantic analysis on the virtual commodity price information text, extract third text content fragments containing the public price information of virtual items, and determine the public price information of virtual items based on the text content about the public price of virtual items in each of the third text content fragments.

[0014] Optionally, determining the estimated transaction price of the current virtual item based on the current valuation reference information includes:

[0015] The first initial estimated price of the current virtual item is determined based on the attribute values ​​of the current virtual item;

[0016] The second initial estimated price of the current virtual item is determined based on the historical transaction price of the current virtual item or the historical transaction price of similar virtual items of the same category or with a category similarity greater than a preset value.

[0017] The third initial estimated price of the current virtual item is determined based on the publicly available price information of the current virtual item;

[0018] The base estimated price of the current virtual item is determined based on the first initial estimated price, the second initial estimated price, and the third initial estimated price.

[0019] Based on the aforementioned basic estimated price, the estimated transaction price is determined according to the current virtual item price discussion information and / or price change trend information.

[0020] Optionally, the verification result includes normal transactions, suspicious transactions, and illegal transactions. When the verification result is a normal transaction, the current virtual item transaction is allowed to complete. When the verification result is an illegal transaction, the current virtual item transaction is canceled. When the verification result is a suspicious transaction, the current virtual item transaction is allowed, and a suspicious count is added to the transaction for both parties. Users whose suspicious count reaches a preset upper limit cannot participate in virtual item transactions.

[0021] Optionally, the user-related information includes the violation history information of each transaction user; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item transaction, and the current user-related information, and the verification result is output, including:

[0022] If the violation history information of either party in the virtual item transaction meets the preset violation transaction conditions, the verification result of the virtual item transaction is output as a violation transaction.

[0023] Optionally, the user-related information includes the login IP information of the trading user; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item, and the current user-related information, and the verification result is output, including:

[0024] Determine whether the difference between the pending transaction price and the estimated transaction price is greater than a preset normal difference value;

[0025] If the difference is not greater than the normal difference value, the verification result of the virtual item transaction is determined to be a normal transaction;

[0026] If the difference is greater than the normal difference value, determine the login IP information of both parties in the transaction. If the login IP information is the same, output the verification result of the virtual item transaction as a suspicious transaction.

[0027] Optionally, the user-related information includes social relationships between the trading users; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item, and the current user-related information, and the verification result is output, including:

[0028] Determine whether the difference between the pending transaction price and the estimated transaction price is greater than a preset normal difference value;

[0029] If the difference is not greater than the normal difference value, the verification result of the virtual item transaction is determined to be a normal transaction;

[0030] If the difference is greater than the normal difference value, determine whether there is a social relationship between the two users. If there is, output the verification result of the virtual item transaction as a normal transaction; if not, output the verification result of the virtual item transaction as an illegal transaction.

[0031] Optionally, the user-related information includes the historical user posts of each trading user; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item, and the current user-related information, and the verification result is output, including:

[0032] Determine whether the difference between the pending transaction price and the estimated transaction price is greater than a preset normal difference value;

[0033] If the difference is not greater than the normal difference value, the verification result of the virtual item transaction is output as a normal transaction;

[0034] If the difference is greater than the normal difference value, the semantic analysis of the historical user statements of both parties in the transaction is performed using the large language model to determine the frequency of the occurrence of preset prohibited words in the historical user statements. If the frequency is greater than the preset value, the verification result of the virtual item transaction is output as an illegal transaction.

[0035] According to another aspect of the present invention, a virtual item transaction data processing apparatus is provided, comprising:

[0036] The information extraction unit is used to collect information text related to historical virtual item transactions, and extract historical valuation reference information and historical user-related information of historical transaction users from the information text through the large language model.

[0037] The estimated transaction price determination unit is used to, when a current virtual item transaction is detected, determine the current valuation reference information related to the current virtual item from the historical valuation reference information, determine the estimated transaction price of the current virtual item based on the current valuation reference information, and determine the current user information of the current trading parties of the current virtual item transaction from the historical user information information.

[0038] The verification execution unit is used to verify the current virtual item transaction based on the estimated transaction price, the pending transaction price of the current virtual item transaction, and the current user's relevant information, and output the obtained verification result.

[0039] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0040] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the virtual item transaction data processing method according to any embodiment of the present invention.

[0041] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the virtual item transaction data processing method according to any embodiment of the present invention.

[0042] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements the virtual item transaction data processing method according to any embodiment of the present invention.

[0043] The technical solution of this invention extracts historical valuation reference information and historical user information of historical transaction users from information text through a large language model, and considers the above factors when making valuations, thereby obtaining a more accurate estimated transaction price of the current virtual item. Based on the estimated transaction price and the price to be traded, the current user information is used for verification to obtain a more accurate verification result, thereby solving the problem of low valuation accuracy in existing virtual item transaction verification, which affects the transaction judgment result.

[0044] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart of a virtual item transaction data processing method provided in Embodiment 1 of the present invention;

[0047] Figure 2 This is a flowchart of a method for extracting historical valuation reference information provided in Embodiment 2 of the present invention;

[0048] Figure 3This is a flowchart of a method for determining the estimated transaction price provided in Embodiment 2 of the present invention;

[0049] Figure 4 This is a schematic diagram of the structure of a virtual item transaction data processing device provided in Embodiment 3 of the present invention;

[0050] Figure 5 This is a schematic diagram of the structure of an electronic device that implements the virtual item transaction data processing method of this invention. Detailed Implementation

[0051] To enable those skilled in the art to better understand the present invention, 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0052] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0053] Example 1

[0054] Figure 1 This is a flowchart of a virtual item transaction data processing method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations involving the verification of virtual item transactions. The method can be executed by a virtual item transaction data processing device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0055] S110. Collect information text related to historical virtual item transactions, and extract historical valuation reference information and historical user-related information of historical transaction users from the information text through a large language model.

[0056] Information text related to historical virtual item transactions includes, but is not limited to, the following categories, taking in-game equipment as an example of virtual items:

[0057] Official historical transaction log text, including the in-game transaction system archive structured log text: Player A (UID123) sold Equipment X (ITEM001) for 5000 gold, Buyer B (UID456), Transaction channel: Auction House.

[0058] Player history transaction dialogue text, including in-game private chat or transaction channel saves, forum history posts, for example, Player A: "I bought equipment X during the 2024 anniversary event for 5500 game coins, now selling it, 4800 game coins negotiable"; Player B: "Based on the transaction at the same time last year, this price is too high, I want it for 4500 game coins.

[0059] Official historical rules texts, including historical version announcements, event descriptions, update logs, etc.; for example: Version update log: 2025-01-01 Version update: Equipment X rarity changed from "Legendary" to "Epic", dungeon drop rate increased from 0.1% to 0.5%.

[0060] Third-party historical transaction texts, including archives from third-party trading platforms, such as: Platform record text: September 2024 Equipment X transaction records: 32 transactions in total, average price 4900 gold, lowest price 4200 game coins (urgent sale), highest price 5800 game coins.

[0061] Historical violation judgment texts are archived in the game compliance system. For example: Judgment text: 2024-09-10, Player C (UID789) sold equipment X (4000 game coins) and was judged as an "abnormal transaction".

[0062] First, various types of textual information related to past virtual item transactions are collected. Then, a large language model is used to perform semantic parsing and information extraction on these unstructured texts, extracting two key types of content: historical information that is valuable for current item valuation, such as the historical transaction prices of various virtual items and factors affecting prices, such as price fluctuations during rarity changes or version adjustments. This type of information serves as a historical stock price reference for virtual items. Second, user-related information that participated in historical transactions, such as user transaction frequency, historical violation records, and activity levels, reflects user behavior characteristics and provides a foundation for subsequent dynamic valuation and user behavior analysis based on historical data.

[0063] S120. When a current virtual item transaction is detected, determine the current valuation reference information related to the current virtual item from the historical valuation reference information, and determine the estimated transaction price of the current virtual item based on the current valuation reference information; determine the current user information of the current trading parties of the current virtual item transaction from the historical user information information.

[0064] When a virtual item transaction is detected, feature matching is first used to filter out highly relevant historical valuation reference information. For example, if the current transaction is for "V3.2 version equipment X", the transaction prices and price fluctuation patterns of the same version or type of weapon during similar events in the past will be extracted, such as the price reduction when supply increases and the key factors affecting the price, such as the premium ratio after version enhancement, as direct references for the current valuation. Then, this relevant historical information is combined with current market dynamics such as real-time supply and demand and version rules to calculate the estimated transaction price of the current virtual item.

[0065] At the same time, by associating user identities, such as account ID matching, the system can extract past data of both parties in the current transaction from relevant information of historical users. This includes, but is not limited to, historical transaction frequency, violation records (such as whether offline transactions are involved), activity level (such as login duration), and transaction preferences (such as whether low-price bulk transactions are frequently conducted). This provides a basis for judging whether there is any risk in the current transaction based on user behavior dimensions.

[0066] S130. Verify the current virtual item transaction based on the estimated transaction price, the pending transaction price of the current virtual item transaction, and the current user's relevant information, and output the verification result.

[0067] First, using the estimated transaction price as a benchmark, calculate the deviation of the pending transaction price. If the pending transaction price is within a reasonable fluctuation range of the estimated price, it is preliminarily determined that there is no abnormality in the price. If the deviation exceeds the threshold (e.g., the pending transaction price of 1000 game coins is much lower than the estimated price of 3600 game coins, a deviation exceeding 70%), then the transaction may be risky. Verify the reasonableness of the risk by combining the user information of both parties in the transaction: If the current seller is an active user with no historical violations, and the user information shows a recent sharp drop in login time, or if the current buyer is a high-paying, low-violation quality user, even if the pending transaction price is slightly lower than the estimated price, the risk may be eliminated. Conversely, if the user information shows a history of illegal item trading or frequent social connections involving item exchanges, even a small price deviation will strengthen the suspicion of illegal transactions. Based on the above analysis, the verification results are output as follows: if the price is normal and the user is risk-free, a normal transaction is output and automatically released; if the price is critically abnormal or the user has a slight risk, a suspicious transaction is output and a review is triggered; if the price is severely abnormal and the user's behavior risk is clear, an illegal transaction is output and the transaction is blocked. At the same time, a natural language report containing the price deviation range, user risk label, and judgment basis can be generated, such as "The pending transaction price of 1000 game coins is 72% lower than the estimated price of 3600 game coins, and the seller is a historically violating alt account, which is judged as a violation and suspected of being a bulk item transfer," which can be used by the operations or compliance team for traceability.

[0068] Example 2

[0069] Figure 2 This is a flowchart of a method for extracting historical valuation reference information provided in Embodiment 2 of the present invention. This embodiment further explains and illustrates the method based on the above embodiments. The information text includes virtual commodity price discussion text, virtual item attribute change text, and virtual commodity price information text obtained from at least one message acquisition source; the historical valuation reference information includes price discussion information, price change trend information, and publicly available price information. Figure 2 As shown, the method includes:

[0070] S210. Perform semantic analysis on the discussion text using a large language model, extract the first text content fragment containing discussion information about the price of virtual items, and determine the discussion information about the price of virtual items based on the text content about the price of virtual items in each first text content fragment.

[0071] Information text represents unstructured text related to transactions, focusing on player-generated discussions and attribute announcements from official or authoritative platforms, providing raw material for historical valuation references. Discussion texts from virtual item trading forums, originating from vertical forums frequented by players, primarily consist of real-time interactions among players regarding virtual item transactions. Examples include "Buying equipment X, up to 4000 game coins," "Selling equipment X from the anniversary sale, 3800 game coins negotiable, just got it last night," and "Tested and found that equipment X's slow effect trigger rate is lower, so it's not fetching a good price now." These texts contain players' price expectations and negotiation demands, as well as implicitly reflect the actual functionality of the items, making them crucial for capturing real market price feedback.

[0072] The attribute change text on virtual item information websites comes from official websites, authoritative strategy platforms, or official partner information sites. The content is mainly an announcement of virtual item attribute adjustments, such as "XX Game V3.2 version update announcement: Equipment X's base attack has been increased from 450 to 500, and the probability of slowing down has been increased from 20% to 25%" or "Anniversary event announcement: From September 1st to 7th, Equipment X can be obtained through event exchange, and its attributes are consistent with those dropped in dungeons." This type of text directly determines the "basic value benchmark" of the item and is an authoritative basis for analyzing the reasons for price changes.

[0073] By performing semantic recognition, entity extraction, and trend analysis on the aforementioned information text using LLM, two types of historical valuation reference information can be generated, directly serving the valuation and verification of current transactions. Price discussion information refers to player price demands and market supply and demand feedback extracted from forum discussion texts. For example, extracting "historical transaction price range (3800-4000 game coins)" from "Buying equipment X up to 4000 game coins" and "Selling equipment X for 3800 gold game coins," and extracting historical supply and demand relationships from "few buyers, many sellers," this type of information can reflect the true market psychological price in historical transactions, avoiding the problem of being out of touch with players' actual understanding caused by relying solely on official pricing or average prices.

[0074] Price trend information refers to the patterns of price changes over time and in different scenarios, analyzed by combining attribute change texts and long-term discussion texts. For example, by combining the attribute change text of "Equipment X Attack Enhancement" in version V3.2 with the prices in forum discussion texts before and after the version update (3500 game coins before the update, 4000 game coins after the update), the trend of "version enhancement leading to a 14% price increase" can be extracted. Similarly, by combining the anniversary return announcement with price discussions during the return period (4500 game coins before the return, 3800 game coins during the return), the trend of "increased supply leading to a 16% price decrease" can be extracted. This type of information can provide a scenario-based basis for current price adjustments. For example, if we are currently in a similar event period, we can refer to historical trends to predict the magnitude of price fluctuations.

[0075] LLM leverages its semantic parsing capabilities to extract textual information and historical valuation references. For example, LLM first crawls attribute change text from virtual item information websites, then crawls discussion text before and after the version update from trading forums. Through semantic matching, it strengthens the association between attributes and price increases, ultimately extracting price trend information, such as "Attack +50 → Price +14%". Simultaneously, it extracts price discussion information from real-time forum discussions. Together, these two elements constitute the item's historical valuation reference information. When trading this item, based on this historical information and the current version or event, the estimated transaction price can be accurately calculated, and the reasonableness of the pending transaction price can be determined.

[0076] First, LLM performs semantic filtering and fragment extraction on the original discussion text: by identifying keywords related to "price" in the text, such as "game currency," "price," "buy," "sell," "buy," "XX yuan," and "negotiable," as well as semantic intent, such as transaction requests, price evaluations, and price negotiations, it filters out the first text content fragments containing discussions about the price of virtual items. For example, from discussion texts such as "Buying V3.2 version equipment X, bring your price, my target price is under 4000 game currency," "Selling equipment X from the anniversary event return, 3800 game currency, no bargaining, just got it last night," and "This equipment X is dropping in price quickly now, it could sell for 4500 last week, now it's hard to even get 4000," three first text content fragments are extracted, all focusing on the price discussion of equipment X. LLM then performs structured price information extraction on each first text content fragment, including the virtual item name, transaction direction, price value, price conditions, time point, and subjective evaluation. For example, from the statement "Selling anniversary return item X, 3800 game coins, no bargaining," we can extract "Item: Equipment X (anniversary return), Direction: Sell, Price: 3800 game coins, Condition: No bargaining"; from the statement "This equipment X is dropping in price quickly, last week it could sell for 4500, now even 4000 is difficult," we can extract "Item: Frost Staff, Time: Last week vs. now, Price change: 4500 game coins to 4000 game coins (price drop), Evaluation: Difficult to sell." LLM determines the mainstream price range (e.g., 3800-4000 game coins) by integrating price values ​​from multiple segments; it judges the supply and demand relationship by combining the direction and quantity of transactions; it extracts price trend evaluations by describing time nodes and price changes; and it clarifies the market's bargaining space by summarizing price conditions.

[0077] S220. Perform semantic analysis on the text of virtual item attribute changes using a large language model, extract second text content segments containing the attribute change values ​​of virtual items, and determine the price change trend information of the virtual item based on the attribute change values ​​of the virtual items in each second text content segment.

[0078] LLM first identifies and locates key terms related to attribute changes in the text of virtual item attribute changes, filtering out second text fragments containing specific attribute value changes. This type of text typically includes core elements such as attribute type (e.g., attack, defense, trigger probability), the values ​​before and after the change, and the direction of the change. LLM accurately locates relevant fragments by recognizing semantic markers such as "+", "-", "from X to Y", and "increase / decrease". For example, from the V3.2 version announcement of the game "XX Game" stating "Equipment Adjustment: Equipment X's base attack increased from 450 to 500, and the slow trigger probability increased from 20% to 25%; simultaneously, the equipment weight decreased from 3.5kg to 3.0kg", LLM extracts three second text fragments: "Base attack increased from 450 to 500"; "Slow trigger probability increased from 20% to 25%"; and "Equipment weight decreased from 3.5kg to 3.0kg". LLM then sorts the extracted second text fragments by attribute importance and performs price impact logic reasoning, combining this with the functional scenarios of the virtual item to transform the attribute changes into quantifiable price change trends. The weight of different attributes on price varies significantly. LLM prioritizes core functional attributes by correlating game common sense with historical data. For example, for "Equipment X" as a weapon, "base attack" and "trigger probability" are core attributes affecting actual combat effectiveness, while weight is a secondary attribute. Therefore, the focus is on analyzing changes in the first two types of attributes. LLM calculates the magnitude of changes in core attributes and combines this with the correlation between historical adjustments to similar attributes and price changes, such as "for every 10 points increase in attack, the price increases by an average of 5%", to deduce the price trend corresponding to the current attribute change.

[0079] S230. Perform semantic analysis on the virtual commodity price information text using a large language model, extract third text content fragments containing the public price information of virtual items, and determine the public price information of the virtual item based on the text content about the public price of the virtual item in each third text content fragment.

[0080] LLM performs semantic analysis on publicly available price information texts for virtual goods, such as official game pricing announcements, listed prices on legitimate trading platforms, and reference prices published by authoritative strategy guides. It precisely filters out third-party text fragments containing explicit publicly available price information. Specifically, LLM identifies keywords such as "official price," "listed price," "reference price," and "sell price," along with their corresponding price values ​​and units, to pinpoint core fragments. For example, it extracts relevant fragments from phrases like "XX game official announcement: anniversary limited skin publicly priced at 1980 points" or "X shopping platform listing shows equipment X publicly referenced price range of 5000-6000 game coins." By integrating publicly available price information from all third-party text fragments and eliminating duplicate or contradictory data, it ultimately determines the publicly available price information for the virtual item. This price information may be a specific value, a fixed range, or an authoritatively labeled price benchmark, serving as a reference for valuation and transaction verification.

[0081] Figure 3 This is a flowchart of a method for determining the estimated transaction price provided in Embodiment 2 of the present invention, as follows: Figure 3 As shown, the method includes:

[0082] S310. Determine the first initial estimated price of the current virtual item based on its attribute values.

[0083] The key attribute values ​​of the current item are determined, such as basic functional attributes: level 60, attack +500, defense +120; rarity values: legendary items correspond to a quantitative indicator of "limited to 1000 pieces across the entire server"; special effect values: probability of triggering a 25% slow effect, strength of adding 15% fire damage, etc. Then, based on preset attribute price association rules, which are generated by analyzing the attributes and transaction prices of similar items in the past (e.g., every 10 points increase in attack corresponds to a price increase of 500 game coins, legendary rarity has a 50% premium over epic rarity, every 5% increase in special trigger probability corresponds to an 8% price increase), each attribute value is converted into a corresponding value component. Through weighted summation, core attributes such as attack have a higher weight than secondary attributes such as equipment weight, to calculate the initial price. This price is a base price based on the item's inherent attributes.

[0084] S320. Determine the second initial estimated price of the current virtual item based on the historical transaction price of the current virtual item or the historical transaction price of similar virtual items of the same category or with a category similarity greater than a preset value.

[0085] If the current virtual item has a history of transactions, the valid data from its historical transaction prices are directly extracted, and the mean, median, or weighted average is calculated to obtain the basic price reference for the item. If the current virtual item is a new product, has very little stock, or has no direct historical transaction records, similar virtual items of the same category or with a category similarity exceeding a preset threshold, such as attribute overlap ≥80%, rarity, or functional scenario consistency, are selected. The historical transaction prices of these similar items are extracted and adjusted for scenario adaptation to finally determine the second initial estimated price of the current item.

[0086] S330. Determine the third initial estimated price of the current virtual item based on the publicly available price information of the current virtual item.

[0087] Based on publicly available price information for virtual items, if there is a clear official price, such as the selling price released by the game's official website or the price for redeeming items during events, this price can be used as the third initial estimated price. If the publicly available price information consists of reference prices or price ranges from multiple authoritative channels, the data is integrated by calculating the mean, median, or by assigning a higher weight to recent publicly available prices. At the same time, publicly available price data with incorrect labeling or abnormal deviations are removed to ensure the accuracy of the results. The resulting third initial estimated price has strong stability and credibility.

[0088] S340. Determine the base estimated price of the current virtual item based on the first initial estimated price, the second initial estimated price, and the third initial estimated price.

[0089] The basic estimated price of a virtual item is determined by dynamically weighting and balancing three initial estimated price dimensions: the first initial estimated price represents the objective value related to the attribute, the second initial estimated price represents the historical transaction value in the market, and the third initial estimated price represents the publicly available value. First, weights are assigned based on the item's context and data sufficiency. If the item is new (no historical transactions or publicly available price), the first and third initial estimated prices are given priority. If the item is a mature category (rich historical transaction data and the publicly available price is only for reference), the second initial estimated price is given priority. If the publicly available price is the only official price, the third initial estimated price has the highest weight. Simultaneously, a large language model is used to verify the consistency of the three price categories, eliminating outliers. Finally, the basic estimated price is calculated through a weighted summation.

[0090] S350. Based on the basic estimated price, determine the estimated transaction price according to the current virtual item price discussion information and / or price change trend information.

[0091] If there is current discussion about the price of virtual items, the base price is adjusted based on this real trading sentiment. If the feedback indicates oversupply, the base price is slightly adjusted downwards; if the feedback indicates undersupply, it is slightly adjusted upwards. If there is price trend information, the price is further adjusted based on the trend magnitude. For example, if it is currently an anniversary sale rerun, the base price is adjusted according to historical rerun price reductions. If both types of information exist simultaneously, the more timely price discussion information is prioritized, and then the long-term patterns of price trend information are superimposed to calculate an estimated trading price that fits the current market reality. This approach retains the objective attributes of the base price while incorporating real-time supply and demand and trend predictions, preventing the valuation from deviating from market dynamics.

[0092] In this embodiment of the invention, the verification result includes normal transaction, suspicious transaction, and illegal transaction. When the verification result is a normal transaction, the current virtual item transaction is allowed to be completed. When the verification result is an illegal transaction, the current virtual item transaction is canceled. When the verification result is a suspicious transaction, the current virtual item transaction is allowed, and a suspicious count is added to the transaction for both parties. Users whose suspicious count reaches a preset upper limit cannot participate in virtual item transactions.

[0093] The verification results are divided into three categories: normal transactions, suspicious transactions, and illegal transactions, each with a different transaction processing strategy: If a transaction is determined to be normal, it means that the estimated transaction price of the virtual item matches the price to be traded, and both users have no risky behavior, so the transaction will be allowed to complete smoothly; if a transaction is determined to be illegal, it means that there is a clear risk in the transaction, and the system will directly cancel the transaction to block the risk; if a transaction is determined to be suspicious, the transaction will be allowed to continue, but suspicious counts will be added to both users. When a user's suspicious count accumulates to a preset limit, that user will be restricted from participating in all subsequent virtual item transactions.

[0094] In this embodiment of the invention, user-related information includes the violation history information of each transaction user; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item, and the current user-related information, and the verification result is output, including:

[0095] If the violation history information of either party in a virtual item transaction meets the preset violation transaction conditions, the verification result of the virtual item transaction will be output as a violation transaction.

[0096] User-related information includes the past violation history of each user participating in virtual item transactions, such as violation records marked for offline transactions, bulk item farming, and zero-price transfers. Based on this, when verifying the current virtual item transaction, the system comprehensively combines the estimated transaction price of the item, the actual pending transaction price, and the user-related information of both parties in the current transaction, and outputs the result according to the verification rules. One key rule is: if the violation history of either the buyer or seller in the current transaction meets the system's preset violation transaction conditions, such as "3 or more violation records in the past 90 days" or "existing serious violation records such as 'item farming' and not yet expired," then there is no need to over-consider the price deviation, and the verification result of the virtual item transaction is directly output as a violation transaction.

[0097] In this embodiment of the invention, user-related information includes the login IP information of the transaction user; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item, and the current user-related information, and the verification result is output, including:

[0098] Determine whether the difference between the pending transaction price and the estimated transaction price is greater than the preset normal difference value;

[0099] If the difference is not greater than the normal value, the verification result of the virtual item transaction is determined to be a normal transaction;

[0100] If the difference is greater than the normal value, determine the login IP information of both parties in the transaction. If the login IP information is the same, output the verification result of the virtual item transaction as a suspicious transaction.

[0101] The user-related information specifically includes the login IP information of the buyer and seller users participating in the virtual item transaction. When verifying the current virtual item transaction, a comprehensive judgment is made by combining the estimated transaction price, the actual pending transaction price, and the login IP information of both parties. The specific verification logic is as follows: First, the difference between the pending transaction price and the estimated transaction price is calculated, and it is determined whether the difference exceeds the preset normal difference value. If the calculated difference does not exceed the normal difference value, it means that the price meets market expectations, and the verification result of the virtual item transaction is directly determined as a normal transaction. If the difference exceeds the normal difference value, the login IP information of both parties will be further retrieved and verified. If it is found that the login IP information of both parties is exactly the same, and there is suspicion that they are operating on the same device or in the same network environment, the verification result of the virtual item transaction will be output as a suspicious transaction.

[0102] In this embodiment of the invention, user-related information includes social relationships between trading users; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item, and the current user-related information, and the verification result is output, including:

[0103] Determine whether the difference between the pending transaction price and the estimated transaction price is greater than the preset normal difference value;

[0104] If the difference is not greater than the normal value, the verification result of the virtual item transaction is determined to be a normal transaction;

[0105] If the difference is greater than the normal value, determine whether there is a social relationship between the two users. If there is, output the verification result of the virtual item transaction as a normal transaction; if not, output the verification result of the virtual item transaction as an illegal transaction.

[0106] User-related information can include social connections between the buyer and seller in virtual item transactions, such as whether they are in-game friends, belong to the same guild, or have a history of teaming up or trading together. When verifying a current virtual item transaction, the following logic is used to comprehensively judge and output the result: First, the difference between the pending transaction price and the estimated transaction price is calculated, and it is determined whether the difference exceeds the preset normal difference value. If the difference does not exceed the normal difference value, the transaction is directly determined to be a normal transaction. If the difference exceeds the normal difference value, the existence of the aforementioned social relationship between the two parties is further verified. If a social relationship is confirmed, it may be a reasonable scenario such as a normal low-price transfer between relatives and friends, and the transaction is still determined to be a normal transaction. If there is no social relationship, it is more likely to be an abnormally low-price transaction between strangers, with risks such as item fraud, and the transaction is determined to be an illegal transaction.

[0107] In this embodiment of the invention, user-related information includes the historical user posts of each trading user; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item, and the current user-related information, and the verification result is output, including:

[0108] Determine whether the difference between the pending transaction price and the estimated transaction price is greater than the preset normal difference value;

[0109] If the difference is not greater than the normal value, the verification result of the virtual item transaction is output as a normal transaction;

[0110] If the difference is greater than the normal difference value, semantic analysis is performed on the historical user statements of both parties in the transaction using a large language model to determine the frequency of the preset prohibited words in the historical user statements. If the frequency is greater than the preset value, the verification result of the virtual item transaction is output as an illegal transaction.

[0111] User-related information can include the past statements of both buyers and sellers involved in virtual item transactions, such as in-game private chats, trade channel announcements, and forum posts. When verifying a current virtual item transaction, the system will make a comprehensive judgment and output the result according to the following logic: First, calculate the difference between the pending transaction price and the estimated transaction price, and determine whether the difference exceeds the preset normal difference value. If the difference does not exceed the normal difference value, the verification result is directly output as a normal transaction. If the difference exceeds the normal difference value, the system will perform semantic analysis on the historical user statements of both parties using a large language model to identify whether they contain preset prohibited words, such as words related to illegal transactions, such as "offline transfer," "trading platform cashback," and "cheating to farm items," and count the frequency of these prohibited words. If the frequency exceeds the preset threshold, such as the word "offline transaction" appearing 5 times or more in a user's historical statements, the verification result is output as an illegal transaction. In this way, the semantic clues in the user's historical statements can help identify possible illegal intentions hidden behind abnormal prices.

[0112] Example 3

[0113] Figure 4 This is a schematic diagram of the structure of a virtual item transaction data processing device provided in Embodiment 3 of the present invention. Figure 4 As shown, the device includes:

[0114] The information extraction unit 410 is used to collect information text related to historical virtual item transactions, and extract historical valuation reference information and historical user-related information of historical transaction users from the information text through a large language model.

[0115] The estimated transaction price determination unit 420 is used to determine the current valuation reference information related to the current virtual item from the historical valuation reference information when the current virtual item transaction is detected, and to determine the estimated transaction price of the current virtual item based on the current valuation reference information; and to determine the current user information of the current trading parties from the historical user information.

[0116] The verification execution unit 430 is used to verify the current virtual item transaction based on the estimated transaction price, the pending transaction price of the current virtual item transaction, and the current user's relevant information, and output the obtained verification result.

[0117] The virtual item transaction data processing device provided in this embodiment of the invention can execute the virtual item transaction data processing method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0118] Example 4

[0119] Figure 5A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, 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 can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), 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 invention described and / or claimed herein.

[0120] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0121] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0122] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 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 11 performs the various methods and processes described above, such as virtual item transaction data processing methods.

[0123] In some embodiments, the virtual item transaction data processing method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the virtual item transaction data processing method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the virtual item transaction data processing method by any other suitable means (e.g., by means of firmware).

[0124] 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.

[0125] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0126] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. 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 thereof.

[0127] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. 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).

[0128] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations 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., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0129] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through 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. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0130] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0131] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. 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 invention should be included within the scope of protection of this invention.

Claims

1. A method for processing virtual item transaction data, characterized in that, include: Collect information text related to historical virtual item transactions, and extract historical valuation reference information and historical user-related information of historical transaction users from the information text using the large language model; When a current virtual item transaction is detected, the current valuation reference information related to the current virtual item is determined from the historical valuation reference information, and the estimated transaction price of the current virtual item is determined based on the current valuation reference information. The current user information of the current parties to the current virtual item transaction is determined from the historical user information. The current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item transaction, and the current user's relevant information, and the verification result is output.

2. The method according to claim 1, characterized in that, The information text includes virtual commodity price discussion text, virtual item attribute change text, and virtual commodity price information text obtained from at least one message acquisition source; the historical valuation reference information includes price discussion information, price change trend information, and publicly available price information; correspondingly, historical valuation reference information is extracted from the information text using the large language model, including: The large language model is used to perform semantic analysis on the discussion text to extract first text content fragments containing virtual item price discussion information. The price discussion information of the virtual item is determined based on the text content about the price of the virtual item in each of the first text content fragments. The large language model is used to perform semantic analysis on the text of the virtual item attribute change, and a second text content segment containing the attribute change value of the virtual item is extracted. The price change trend information of the virtual item is determined based on the virtual item attribute change value in each second text content segment. The large language model is used to perform semantic analysis on the virtual commodity price information text, extract third text content fragments containing the public price information of virtual items, and determine the public price information of virtual items based on the text content about the public price of virtual items in each of the third text content fragments.

3. The method according to claim 2, characterized in that, Determining the estimated transaction price of the current virtual item based on the current valuation reference information includes: The first initial estimated price of the current virtual item is determined based on the attribute values ​​of the current virtual item; The second initial estimated price of the current virtual item is determined based on the historical transaction price of the current virtual item or the historical transaction price of similar virtual items of the same category or with a category similarity greater than a preset value. The third initial estimated price of the current virtual item is determined based on the publicly available price information of the current virtual item; The base estimated price of the current virtual item is determined based on the first initial estimated price, the second initial estimated price, and the third initial estimated price; Based on the aforementioned basic estimated price, the estimated transaction price is determined according to the current virtual item price discussion information and / or price change trend information.

4. The method according to claim 1, characterized in that, The verification results include normal transactions, suspicious transactions, and illegal transactions. When the verification result is a normal transaction, the current virtual item transaction is allowed to complete. When the verification result is an illegal transaction, the current virtual item transaction is canceled. When the verification result is a suspicious transaction, the current virtual item transaction is allowed, and a suspicious count is added to the transaction for both users. Users whose suspicious count reaches a preset upper limit cannot participate in virtual item transactions.

5. The method according to claim 4, characterized in that, The user-related information includes the violation history information of each transaction user; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item transaction, and the current user-related information, and the verification results are output, including: If the violation history information of either party in the virtual item transaction meets the preset violation transaction conditions, the verification result of the virtual item transaction is output as a violation transaction.

6. The method according to claim 4, characterized in that, The user-related information includes the login IP information of the trading user; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item, and the current user-related information, and the verification result is output, including: Determine whether the difference between the pending transaction price and the estimated transaction price is greater than a preset normal difference value; If the difference is not greater than the normal difference value, the verification result of the virtual item transaction is determined to be a normal transaction; If the difference is greater than the normal difference value, determine the login IP information of both parties in the transaction. If the login IP information is the same, output the verification result of the virtual item transaction as a suspicious transaction.

7. The method according to claim 4, characterized in that, The user-related information includes the social relationships between the trading users; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item, and the current user-related information, and the verification results are output, including: Determine whether the difference between the pending transaction price and the estimated transaction price is greater than a preset normal difference value; If the difference is not greater than the normal difference value, the verification result of the virtual item transaction is determined to be a normal transaction; If the difference is greater than the normal difference value, determine whether there is a social relationship between the two users. If there is, output the verification result of the virtual item transaction as a normal transaction; if not, output the verification result of the virtual item transaction as an illegal transaction.

8. The method according to claim 4, characterized in that, The user-related information includes the historical user posts of each trading user; correspondingly, the current virtual item transaction is verified based on the estimated transaction price, the pending transaction price of the current virtual item, and the current user-related information, and the verification results are output, including: Determine whether the difference between the pending transaction price and the estimated transaction price is greater than a preset normal difference value; If the difference is not greater than the normal difference value, the verification result of the virtual item transaction is output as a normal transaction; If the difference is greater than the normal difference value, the semantic analysis of the historical user statements of both parties in the transaction is performed using the large language model to determine the frequency of the occurrence of preset prohibited words in the historical user statements. If the frequency is greater than the preset value, the verification result of the virtual item transaction is output as an illegal transaction.

9. A virtual item transaction data processing device, characterized in that, include: The information extraction unit is used to collect information text related to historical virtual item transactions, and extract historical valuation reference information and historical user-related information of historical transaction users from the information text through the large language model. The estimated transaction price determination unit is used to determine the current valuation reference information related to the current virtual item from the historical valuation reference information when a current virtual item transaction is detected, and to determine the estimated transaction price of the current virtual item based on the current valuation reference information; The current user information of the current parties to the current virtual item transaction is determined from the historical user information. The verification execution unit is used to verify the current virtual item transaction based on the estimated transaction price, the pending transaction price of the current virtual item transaction, and the current user's relevant information, and output the obtained verification result.

10. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the virtual item transaction data processing method according to any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the virtual item transaction data processing method according to any one of claims 1-8.

12. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the virtual item transaction data processing method according to any one of claims 1-8.

Images