An e-commerce platform consumption risk assessment method

Abstract

The application relates to the technical field of fraud identification, and discloses an e-commerce platform consumption risk assessment method, which comprises the following steps: obtaining target information, forming analysis data through an information analysis strategy, forming judgment data through a seller judgment strategy, evaluating the seller through an evaluation strategy to form evaluation data, warning the buyer and protecting the fund, the e-commerce platform consumption risk assessment method can automatically obtain the risk condition of the seller corresponding to the goods according to the goods purchased by the buyer, automatically judge the authenticity of the goods comment according to the comment text of the goods of the seller, thereby judging whether the seller and the goods have risks, automatically judging whether the seller is normal sales according to the IP of the seller, and prompting the buyer that the present consumption has risks, so that the risk of being cheated by the buyer is reduced, and the property loss caused by the buyer is prevented.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fraud identification, in particular to an e-commerce platform consumption risk assessment method. BACKGROUND

[0002] The e-commerce platform consumption risk assessment method can ensure that consumers on the e-commerce platform can shop and transact more safely and confidently during the transaction process, effectively reduce the risk perception of e-commerce platform consumers, improve the credibility and competitiveness of the platform, and promote the healthy development of the e-commerce industry, including a reputation rating system, payment security measures, consumer protection policies, data encryption and privacy protection, risk warning mechanisms, consumer education and guidelines, and technical audits and certifications.

[0003] Establish and maintain a reputation rating system for consumers and merchants, based on historical transaction records, evaluations and complaints, which can help consumers judge the credibility and service quality of merchants, use secure payment platforms and third-party payment systems to ensure the safety of consumers' funds during the transaction process, develop clear consumer protection policies, including return and exchange policies, complaint handling procedures, etc., to protect consumer rights and interests and reduce consumer risk, strengthen the encryption storage and transmission of user data to ensure that personal privacy information is not leaked, prevent information from being used by criminals, establish a risk warning mechanism to monitor transaction behavior and patterns, and timely identify and respond to potential fraud and risk behavior, provide consumer education and shopping guidelines to help consumers identify and avoid risks in transactions, and enhance consumers' self-protection awareness, conduct technical audits and certifications on merchants and products to ensure that they meet relevant regulations and quality standards, and reduce the risk of consumer selection and purchase.

[0004] The existing e-commerce platform consumption risk assessment method cannot automatically obtain the risk situation of the seller corresponding to the goods purchased by the buyer according to the goods, cannot automatically judge the authenticity of the goods comment according to the comment text of the goods of the seller, and cannot automatically judge whether the seller is normal sales according to the IP of the seller, and prompt the buyer that the current consumption has risks. Due to the limited message channel, the buyer is difficult to judge whether the seller has fraudulent behavior, thereby causing property loss, and the practicality has certain limitations. SUMMARY

[0005] The present application provides an e-commerce platform consumption risk assessment method for promoting the solution to the problems in the background art.

[0006] The present application provides the following technical scheme: an e-commerce platform consumption risk assessment method, comprising:

[0007] Obtaining target information;

[0008] The target information includes seller information and commodity information purchased by the buyer;

[0009] The seller information includes commodity, comment information and equipment information of the seller;

[0010] According to the target information, analysis data is formed through an information analysis strategy to determine whether the comment of the commodity purchased by the buyer and other commodities of the seller of the commodity is a valid comment, so as to determine whether the commodity purchased by the buyer and the seller of the commodity exist risks;

[0011] According to the analysis data, determination data is formed through a seller determination strategy to further determine the commodity and the seller of the commodity which are preliminarily determined to have high risks, so as to prevent misjudgment;

[0012] According to the determination data, the seller is evaluated through an evaluation strategy to form evaluation data, so as to determine whether the commodity or the seller has problems according to the further determination result, thereby warning the buyer and protecting the fund;

[0013] According to the evaluation data, when the commodity purchased by the buyer or the seller of the commodity has problems, the buyer is warned and the fund is protected.

[0014] As an optional solution of the e-commerce platform consumption risk evaluation method, the information analysis strategy is specifically:

[0015] The commodity confirmed by the buyer is determined as a target commodity;

[0016] The seller corresponding to the target commodity is determined as a target seller;

[0017] All commodities of the target seller are determined as target seller commodities;

[0018] The comment information of each target seller commodity is obtained, and the comment information includes a comment score and comment text;

[0019] The total comment score and the effective score are obtained;

[0020] The effective score is less than 80% of the total comment score;

[0021] The target seller commodity with a comment score greater than or equal to 80% of the total comment score is determined as a good comment commodity;

[0022] The target seller commodity with a comment score greater than the effective score and less than 80% of the total comment score is determined as an effective commodity;

[0023] The number of target seller commodities, good comment commodities and effective commodities is respectively determined as a total number of sellers, a good comment number and an effective number;

[0024] If the number of positive reviews is greater than or equal to 80% of the total number of sellers, then the review judgment strategy will be applied to each positive review product.

[0025] Obtain the quantity of target seller's goods that are determined to have low risk, and define it as a low-risk quantity;

[0026] If the low-risk quantity is greater than or equal to the effective quantity, then the target seller is considered to have low risk.

[0027] If the low-risk quantity is less than the effective quantity, the target seller is considered to be at high risk, and the target commodity determination strategy is implemented.

[0028] If the number of positive reviews is less than 80% of the total number of sellers and the number of positive reviews is greater than or equal to the number of valid reviews, then the target seller is considered to have low risk.

[0029] If the number of positive reviews is less than the number of valid reviews, the target seller is considered to be of high risk, and the target product judgment strategy will be implemented.

[0030] As an optional solution to the e-commerce platform consumer risk assessment method of the present invention, the target product determination strategy specifically includes:

[0031] Obtain the review score of the target product and set it as the target score;

[0032] If the target score is greater than or equal to 80% of the total review score, then the review judgment strategy will be applied to the target product.

[0033] If the target score is less than 80% of the total review score and the target score is greater than or equal to the valid score, then the target product is considered to have low risk.

[0034] If the target score is less than the effective score, the target product is considered to be high-risk, and the seller's judgment strategy will be implemented.

[0035] As an optional solution to the e-commerce platform consumer risk assessment method of the present invention, the comment judgment strategy specifically includes:

[0036] Get all the review text for the product;

[0037] Set comment keywords;

[0038] Comment text containing keywords related to "comment" will be considered invalid.

[0039] The number of product review comments is defined as the number of reviews.

[0040] The number of invalid characters retrieved for a product is defined as the invalid quantity.

[0041] If the number of invalid reviews is greater than or equal to 80% of the total number of reviews, it indicates that the product has few valid reviews and is therefore considered to be of high risk.

[0042] If (number of comments-number of invalid comments) is greater than or equal to 80% of the number of comments, it indicates that the product has many valid comments, and the product is determined to be low risk.

[0043] As an optional solution of the e-commerce platform consumption risk assessment method, the seller determination strategy is specifically:

[0044] The device information of the seller is obtained, and the device information includes the seller device IP and the number of seller devices.

[0045] If the number of seller devices is 1 and the number of seller device IPs is 1, the network IP connected by the seller device IP is obtained, and the target IP is determined.

[0046] If the number of seller devices is 1 and the number of seller device IPs is greater than 1, the number of uses of each seller device IP is obtained, the seller device IP corresponding to the maximum number of uses is extracted, and the common IP is determined.

[0047] The network IP connected by the common IP is obtained, and the target IP is determined.

[0048] If the number of seller devices is greater than 1 and the number of seller device IPs is 1, the network IP connected by the seller device IP is obtained, and the target IP is determined.

[0049] If the number of seller devices is greater than 1 and the number of seller device IPs is greater than 1, the network IP connected by each seller device IP corresponding to each seller device is obtained, and the target IP is determined.

[0050] The risk determination strategy is executed on the target IP.

[0051] As an optional solution of the e-commerce platform consumption risk assessment method, the risk determination strategy is specifically:

[0052] All device IPs connected by the target IP are obtained and determined as judgment IPs.

[0053] The risk level of each judgment IP is obtained, and the risk level includes high risk and low risk.

[0054] If there is a high-risk judgment IP, it is determined that the judgment IP and the target IP are abnormal.

[0055] If all judgment IPs corresponding to the target IP are low risk, it is determined that the target IP is normal.

[0056] If the target IP is abnormal, it is determined that the seller device IP is abnormal, and a false judgment analysis strategy is executed.

[0057] If all target IPs are normal, it is determined that the seller device IP is normal.

[0058] As an optional solution of the e-commerce platform consumption risk assessment method, the misjudgment analysis strategy is specifically:

[0059] All judgment IPs corresponding to the target IP are obtained;

[0060] The number of uses of each judgment IP to the target IP is obtained;

[0061] According to the value of the number of uses, from large to small, the judgment IP corresponding to the largest number of uses is determined as the comparison IP;

[0062] The abnormal judgment IP is obtained and is determined as the abnormal IP;

[0063] The seller device IP is obtained;

[0064] If the comparison IP and the abnormal IP are consistent, the time period of the seller device IP and the abnormal IP using the target IP is obtained respectively, and is respectively determined as the seller use time period and the abnormal use time period;

[0065] If the seller use time period and the abnormal use time period have high coincidence degree, it is determined that the seller device IP is abnormal, and it is determined that the target seller risk is high;

[0066] If the seller use time period and the abnormal use time period have low coincidence degree, it is determined that the seller device IP is normal, and it is determined that the target seller risk is low;

[0067] If the comparison IP and the abnormal IP are inconsistent, the time period determination strategy is executed.

[0068] As an optional solution of the e-commerce platform consumption risk assessment method, the time period determination strategy is specifically:

[0069] The seller use time period and the abnormal use time period are obtained respectively;

[0070] The time period of the comparison IP using the target IP is obtained and is determined as the comparison use time period;

[0071] If the comparison use time period and the abnormal use time period have high coincidence degree, it is determined that the comparison IP is abnormal;

[0072] If the seller use time period and the abnormal use time period have high coincidence degree, or the seller use time period and the comparison use time period have high coincidence degree, it is determined that the seller device IP is abnormal, and it is determined that the target seller risk is high;

[0073] If the seller use time period and the abnormal use time period have low coincidence degree, and the seller use time period and the comparison use time period have low coincidence degree, it is determined that the seller device IP is normal, and it is determined that the target seller risk is low.

[0074] As an optional solution of the e-commerce platform consumption risk assessment method, the evaluation strategy is specifically:

[0075] The determination results of the target seller and the target commodity are obtained respectively.

[0076] If it is determined that the target commodity risk is low, the payment interface is entered.

[0077] If it is determined that the target seller risk is low, the payment interface is entered.

[0078] If it is determined that the target commodity risk is high, the buyer is prompted that the purchased commodity has a risk, and the buyer is suggested to cancel the order.

[0079] If it is determined that the target seller risk is high, the buyer is prompted that the commodity corresponding to the merchant has a risk, and the buyer is suggested to cancel the order.

[0080] If the buyer chooses to continue to pay, the funds in the target seller account are locked after the buyer pays, and the funds are unlocked after the buyer confirms that the consumption is correct.

[0081] If the buyer provides proof within the time of locking the funds, the funds are directly transferred back to the payment account of the buyer.

[0082] The present application has the following advantages:

[0083] 1. The e-commerce platform consumption risk assessment method obtains the comment information of the target commodity confirmed by the buyer, obtains the comment information of the target commodity corresponding to the merchant and all commodities of the merchant, compares the comment information with the set keywords, automatically judges whether the comment of the commodity is effective, that is, whether the comment of the commodity belongs to the system automatic praise or the professional comment employed by the merchant, or whether it belongs to the effective comment published by the buyer after normal consumption according to the use condition and the condition of the commodity itself, automatically judges whether the target commodity is a false sale for a fraudulent buyer, such as a concert ticket, and prompts the buyer that the current consumption has a risk, reduces the risk of being cheated by the buyer, and prevents the buyer from causing property loss.

[0084] 2. The e-commerce platform consumption risk assessment method obtains the device IP of the seller, obtains the network IP commonly used by the seller, obtains all device IPs connected by the network IP, obtains the risk degree of each device IP, that is, whether the device has a fraudulent behavior, automatically judges the risk degree of the seller, if there is a high-risk device IP, it is determined as an abnormal device, and the risk degree of the device IP of the seller is increased, so as to automatically judge whether the seller is a normal sale, and prompt the buyer that the current consumption has a risk, reduce the risk of being cheated by the buyer, and prevent the buyer from causing property loss.

[0085] 3、The e-commerce platform consumption risk assessment method obtains the device with the most network IP, identifies it as a comparison device, obtains the time period of network use of the seller device, the abnormal device and the comparison device respectively, and determines the network risk. If the abnormal device and the comparison device have high coincidence in network use, it is determined that the network risk is high. If the abnormal device and the comparison device have low coincidence in network use, it is determined that the network risk is low, indicating that the abnormal device may have a network sharing behavior. If the seller device and the abnormal device have high coincidence in network use, it is determined that the seller risk is high. If the seller device and the abnormal device have low coincidence in network use, it is determined that the seller risk is low, indicating that the seller device may have a network sharing behavior. According to the risk determination result of the seller, it is automatically judged whether the seller is a normal sale, and the buyer is prompted that there is a risk in this consumption, the risk of being cheated by the buyer is reduced, the property loss caused by the buyer is prevented, and at the same time, the seller is prevented from being misjudged, so as to cause certain economic loss. BRIEF DESCRIPTION OF DRAWINGS

[0086] Figure 1 The flowchart of the e-commerce platform consumption risk assessment method of the present application is shown in the figure.

[0087] Figure 2 The correlation diagram of the present application is shown in the figure. DETAILED DESCRIPTION

[0088] The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0089] Embodiment one, an e-commerce platform consumption risk assessment method, referring to Figure 1 , comprising:

[0090] Obtain target information;

[0091] The target information includes seller information and commodity information purchased by the buyer;

[0092] The seller information includes the seller's commodity, comment information and device information;

[0093] According to the target information, the analysis data is formed by information analysis strategy, to judge whether the comment of the commodity purchased by the buyer and the other commodities of the seller of the commodity is effective comment, so as to judge whether the commodity purchased by the buyer and the seller of the commodity exist risk;

[0094] According to the analysis data, the seller judgment strategy is used to form judgment data to further judge the goods with high risk in the preliminary judgment and the seller of the goods, so as to prevent misjudgment;

[0095] According to the judgment data, the seller is evaluated by the evaluation strategy to form evaluation data, so as to determine whether the goods or the seller has a problem according to the further judgment result, thereby warning the buyer and protecting the fund;

[0096] According to the evaluation data, when the goods purchased by the buyer or the seller of the goods has a problem, the buyer is warned and the fund is protected.

[0097] Through the above method, the risk of the seller corresponding to the goods purchased by the buyer is automatically obtained according to the goods purchased by the buyer, the authenticity of the goods comment is automatically judged according to the comment text of the goods of the seller, so as to judge whether the seller and the goods have a risk, the seller is automatically judged according to the IP of the seller, and the buyer is prompted that the current consumption has a risk, so as to reduce the risk of being cheated by the buyer and prevent the buyer from causing property loss.

[0098] Embodiment two, this embodiment is improved on the basis of embodiment one, the e-commerce platform consumption risk evaluation method, the information analysis strategy, specifically:

[0099] The goods confirmed by the buyer are determined as target goods;

[0100] The seller corresponding to the target goods is determined as a target seller;

[0101] All goods of the target seller are determined as target seller goods;

[0102] The comment information of each target seller goods is obtained, the comment information includes comment score and comment text, and the comment score is the overall comment score of the goods, that is, comment score = score evaluated by each buyer who purchases the goods ÷ total number of buyers who purchase the goods;

[0103] The total comment score and the effective score are obtained, the total comment score is specifically 100, and the effective score is 60;

[0104] Among them, the effective score < the total comment score × 80%;

[0105] The target seller goods with a comment score ≥ the total comment score × 80% are determined as good comment goods;

[0106] The target seller goods with a comment score ≥ the effective score and a comment score < the total comment score × 80% are determined as effective goods;

[0107] The quantity of target seller goods, the quantity of good comment goods and the quantity of effective goods are obtained respectively, and are respectively defined as the total quantity of seller, the quantity of good comment and the quantity of effective;

[0108] If the quantity of good comment is greater than or equal to 80% of the total quantity of seller, the comment judgment strategy is executed for each good comment;

[0109] The quantity of target seller goods judged as low risk is obtained, and is defined as the quantity of low risk;

[0110] If the quantity of low risk is greater than or equal to the quantity of effective, the target seller is judged as low risk;

[0111] If the quantity of low risk is less than the quantity of effective, the target seller is judged as high risk, and the target good judgment strategy is executed;

[0112] If the quantity of good comment is less than 80% of the total quantity of seller and the quantity of good comment is greater than or equal to the quantity of effective, the target seller is judged as low risk;

[0113] If the quantity of good comment is less than the quantity of effective, the target seller is judged as high risk, and the target good judgment strategy is executed.

[0114] The target good judgment strategy is specifically:

[0115] The comment score of the target good is obtained, and is defined as the target score;

[0116] If the target score is greater than or equal to 80% of the total comment score, the comment judgment strategy is executed for the target good;

[0117] If the target score is less than 80% of the total comment score and the target score is greater than or equal to the effective score, the target good is judged as low risk;

[0118] If the target score is less than the effective score, the target good is judged as high risk, and the seller judgment strategy is executed.

[0119] The comment judgment strategy is specifically:

[0120] All comment texts of the goods are obtained, the comment texts are all comments of each good published by all buyers after purchasing the goods according to their own use and the situation of the goods, and the comments have a certain guiding nature for the buyers who want to purchase the goods, so that according to the comment keywords, it is judged whether the comment belongs to the system set default comment or the professional comment paid by the seller, or the effective comment published by the buyer after normal consumption according to their own use and the situation of the goods, wherein the normal consumption is a complete consumption behavior of the buyer after spending the amount of the goods to receive the goods;

[0121] Setting the comment keyword, the comment keyword is the system default comment or the common words of the professional comment hired by the seller, such as the system default praise, the praise of fast logistics speed, etc.

[0122] The comment text containing the comment keyword is identified as invalid text;

[0123] Get the number of comment text of the product, and set it as the comment number;

[0124] Get the number of invalid text of the product, and set it as the invalid number;

[0125] If the invalid number ≥ comment number × 80%, it means that the valid comment of the product is less, and the risk of the product is high;

[0126] If (comment number - invalid number) ≥ comment number × 80%, it means that the valid comment of the product is more, and the risk of the product is low.

[0127] Embodiment three, this embodiment is improved on the basis of embodiment two, refer to Figure 2 In this embodiment, the seller judgment strategy is specifically:

[0128] Get the device information of the seller, including the seller device IP and the number of seller devices;

[0129] If the number of seller devices = 1, and the number of seller device IP = 1, get the network IP connected by the seller device IP, and identify it as the target IP;

[0130] If the number of seller devices = 1, and the number of seller device IP > 1, get the use frequency of each seller device IP, extract the seller device IP corresponding to the maximum use frequency, and set it as the common IP;

[0131] Get the network IP connected by the common IP, and identify it as the target IP;

[0132] If the number of seller devices > 1, and the number of seller device IP = 1, get the network IP connected by the seller device IP, and identify it as the target IP;

[0133] If the number of seller devices > 1, and the number of seller device IP > 1, get the network IP connected by each seller device IP corresponding to each seller device, and identify it as the target IP respectively;

[0134] Execute the risk judgment strategy on the target IP.

[0135] Among them, the risk judgment strategy is specifically:

[0136] Get all the device IP connected by the target IP, and set it as the judgment IP;

[0137] obtaining a risk level of each determined IP, the risk level including a high risk and a low risk, wherein the high risk is a case where the device IP has once existed fraud, and the low risk is a case where the device IP has temporarily no fraud;

[0138] if there is a determined IP with a high risk, it is determined that the determined IP and the target IP are abnormal;

[0139] if all determined IPs corresponding to the target IP are low risk, it is determined that the target IP is normal;

[0140] if the target IP is abnormal, it is determined that the seller device IP is abnormal, and a false positive analysis strategy is executed;

[0141] if all target IPs are normal, it is determined that the seller device IP is normal.

[0142] The false positive analysis strategy is specifically:

[0143] obtaining all determined IPs corresponding to the target IP;

[0144] obtaining the use frequency of each determined IP to the target IP;

[0145] According to the numerical value of the use frequency, from large to small, the determined IP corresponding to the largest use frequency is determined as a comparison IP;

[0146] obtaining an abnormal determined IP as an abnormal IP;

[0147] obtaining a seller device IP;

[0148] if the comparison IP is consistent with the abnormal IP, the time period of the seller device IP and the abnormal IP using the target IP is obtained respectively, and the seller use time period and the abnormal use time period are determined respectively;

[0149] if the seller use time period and the abnormal use time period have high coincidence degree, it is determined that the seller device IP is abnormal, and it is determined that the target seller risk is high;

[0150] if the seller use time period and the abnormal use time period have low coincidence degree, it is determined that the seller device IP is normal, and it is determined that the target seller risk is low;

[0151] if the comparison IP is inconsistent with the abnormal IP, a time period determination strategy is executed;

[0152] The time period determination strategy is specifically:

[0153] obtaining the seller use time period and the abnormal use time period respectively;

[0154] obtaining the time period of the comparison IP using the target IP as a comparison use time period;

[0155] If the comparison use period and the abnormal use period coincide to a high degree, it is determined that the comparison IP is abnormal;

[0156] If the seller use period and the abnormal use period coincide to a high degree, or the seller use period and the comparison use period coincide to a high degree, it is determined that the seller device IP is abnormal, and it is determined that the target seller is high-risk;

[0157] If the seller use period and the abnormal use period coincide to a low degree, and the seller use period and the comparison use period coincide to a low degree, it is determined that the seller device IP is normal, and it is determined that the target seller is low-risk.

[0158] The embodiment also provides that the evaluation strategy specifically includes:

[0159] The determination results of the target seller and the target commodity are obtained respectively;

[0160] If it is determined that the target commodity is low-risk, a payment interface is entered;

[0161] If it is determined that the target seller is low-risk, a payment interface is entered;

[0162] If it is determined that the target commodity is high-risk, the buyer is prompted that the commodity purchased by the buyer is risky, i.e., there is a possibility of fraud, and the buyer is suggested to cancel the order;

[0163] If it is determined that the target seller is high-risk, the buyer is prompted that the merchant corresponding to the commodity purchased by the buyer is risky, i.e., there is a possibility of fraud, and the buyer is suggested to cancel the order;

[0164] If the buyer chooses to continue to pay, the target seller account is locked after the buyer pays, and the funds are unlocked after the buyer confirms that the consumption is correct;

[0165] If the buyer provides proof that the consumption has a problem within the time of locking the funds, the funds are directly transferred back to the payment account of the buyer.

[0166] In the embodiment, the data of the seller device IP, the target IP, the determination IP, the abnormal IP and the comparison IP are formed into an association graph, the risk of the seller corresponding to the commodity purchased by the buyer is automatically obtained, the authenticity of the commodity comment is automatically judged according to the comment text of the commodity of the seller, whether the seller and the commodity are risky is judged, whether the seller is normal is automatically judged according to the IP of the seller, the buyer is prompted that the consumption has a risk, the risk of being cheated by the buyer is reduced, and the property loss of the buyer is prevented.

[0167] It is to be noted that, as used in this document, the term "indicia" is intended to encompass any type of data, information, or other content, whether in the form of text, graphics, images, video, audio, or otherwise. It is to be further noted that, as used in this document, the terms "coupled" and "connected", along with derivatives thereof, can be used to mean one or more of the following: in electrical communication with; physically touching; in both electrical and physical contact with; and, not in contact with. It is to be further noted that, as used in this document, the terms "include" and "comprise", along with derivatives thereof, can be used to indicate inclusion of one or more elements or steps; these terms are not intended to, nor do they, imply that any or all of the elements or steps are essential permanently, let alone necessarily, combined with any other elements or steps. It is to be further noted that, as used in this document, the term "or" as used in a list of items prefaced by "comprising" or "including" to indicate a disjunctive list of elements, i.e. a list in which items can be selected from the group comprising elements in the list. It is to be further noted that, as used in this document, the terms "first", "second", and the like, merely mean different instances and do not require or imply any actual relationship or order between the referred to items. It is to be further noted that, as used in this document, the terms "include", "includes", and "including" are intended to be non-exclusive such that a process, method, article, or apparatus that includes items recited in the list of items following these terms is not limited to the items in the list, and can include other items not expressly listed or inherent to such process, method, article, or apparatus.

[0168] The above description is merely that of preferred embodiments of the application, and modifications and improvements made to the application in light thereof which are obvious to one of ordinary skill in the art are to be included within the purview of this application.

Claims

1. An e-commerce platform consumption risk assessment method, characterized in that: The application relates to a method for judging the risk of a seller and a buyer, and a system thereof. The method comprises the following steps: acquiring target information; the target information comprises seller information and commodity information purchased by a buyer; the seller information comprises commodities, comment information and equipment information of the seller; according to the target information, analysis data is formed by an information analysis strategy to determine whether the comment of the commodity purchased by the buyer and other commodities of the seller of the commodity is a valid comment, so as to determine whether the commodity purchased by the buyer and the seller of the commodity exist risks; according to the analysis data, judgment data is formed by a seller judgment strategy to further determine the commodity and the seller of the commodity which are preliminarily determined to have high risks, so as to prevent misjudgment; according to the judgment data, the seller is evaluated by an evaluation strategy to form evaluation data, so as to determine whether the commodity or the seller has problems according to the further determination result, thereby warning the buyer and protecting the fund of the buyer; when the commodity purchased by the buyer or the seller of the commodity has problems, the buyer is warned and the fund of the buyer is protected according to the evaluation data; the information analysis strategy is specifically as follows: a commodity confirmed by the buyer is determined as a target commodity; a seller corresponding to the target commodity is determined as a target seller; all commodities of the target seller are determined as target seller commodities; comment information of each target seller commodity is acquired, and the comment information comprises a comment score and comment words; a total comment score and a valid score are acquired; the valid score < the total comment score * 80%; a target seller commodity with a comment score >= the total comment score * 80% is determined as a good comment commodity; a target seller commodity with a comment score >= the valid score and < the total comment score * 80% is determined as a valid commodity; the number of the target seller commodities, the good comment commodities and the valid commodities is respectively determined as a total number of sellers, a good comment number and a valid number; if the good comment number >= the total number of sellers * 80%, a comment judgment strategy is executed on each good comment commodity; the number of target seller commodities determined to have low risks is determined as a low risk number; if the low risk number >= the valid number, the target seller is determined to have low risks; if the low risk number < the valid number, the target seller is determined to have high risks, and a target commodity judgment strategy is executed; if the good comment number < the total number of sellers * 80% and >= the valid number, the target seller is determined to have low risks; if the good comment number < the valid number, the target seller is determined to have high risks, and the target commodity judgment strategy is executed; the target commodity judgment strategy is specifically as follows: a comment score of the target commodity is determined as a target score; if the target score >= the total comment score * 80%, a comment judgment strategy is executed on the target commodity; if the target score < the total comment score * 80% and >= the valid score, the target commodity is determined to have low risks; if the target score < the valid score, the target commodity is determined to have high risks, and a seller judgment strategy is executed; the comment judgment strategy is specifically as follows: all comment words of the commodity are acquired; comment keywords are set; comment words containing the comment keywords are determined as invalid words; the number of the comment words of the commodity is determined as a comment number; the number of the invalid words of the commodity is determined as an invalid number; If the invalid quantity is greater than or equal to 80% of the comment quantity, it indicates that the valid comments of the commodity are few, and it is determined that the commodity risk is high; If (the comment quantity - the invalid quantity) is greater than or equal to 80% of the comment quantity, it indicates that the valid comments of the commodity are many, and it is determined that the commodity risk is low; The seller determination strategy is specifically: Obtain the device information of the seller, including the seller device IP and the number of seller devices; If the number of seller devices = 1, and the number of seller device IPs = 1, obtain the network IP connected by the seller device IP, and determine it as the target IP; If the number of seller devices = 1, and the number of seller device IPs > 1, obtain the use frequency of each seller device IP, extract the seller device IP corresponding to the maximum use frequency, and determine it as the commonly used IP; Obtain the network IP connected by the commonly used IP, and determine it as the target IP; If the number of seller devices > 1, and the number of seller device IPs = 1, obtain the network IP connected by the seller device IP, and determine it as the target IP; If the number of seller devices > 1, and the number of seller device IPs > 1, obtain the network IP connected by each seller device IP corresponding to each seller device, respectively, and determine it as the target IP; Execute the risk determination strategy on the target IP; The risk determination strategy is specifically: Obtain all device IPs connected by the target IP, and determine them as judgment IPs; Obtain the risk level of each judgment IP, including high risk and low risk; If there is a high-risk judgment IP, it is determined that the judgment IP and the target IP are abnormal; If all judgment IPs corresponding to the target IP are low-risk, it is determined that the target IP is normal; If the target IP is abnormal, it is determined that the seller device IP is abnormal, and the misjudgment analysis strategy is executed; If all target IPs are normal, it is determined that the seller device IP is normal; The misjudgment analysis strategy is specifically: Obtain all judgment IPs corresponding to the target IP; Obtain the use frequency of each judgment IP to the target IP; According to the value of the use frequency, sort them from large to small, and determine the judgment IP corresponding to the maximum use frequency as the comparison IP; Obtain the abnormal judgment IP, and determine it as the abnormal IP; Obtain the seller device IP; If the comparison IP is consistent with the abnormal IP, obtain the time period when the seller device IP and the abnormal IP use the target IP, respectively, and determine them as the seller use time period and the abnormal use time period; If the seller use time period and the abnormal use time period have high coincidence degree, it is determined that the seller device IP is abnormal, and it is determined that the target seller risk is high; If the seller use time period and the abnormal use time period have low coincidence degree, it is determined that the seller device IP is normal, and it is determined that the target seller risk is low; If the comparison IP is inconsistent with the abnormal IP, execute the time period determination strategy; The time period determination strategy is specifically: Obtain the seller use time period and the abnormal use time period, respectively; Obtain the time period when the comparison IP uses the target IP, and determine it as the comparison use time period; If the comparison use time period and the abnormal use time period have high coincidence degree, it is determined that the comparison IP is abnormal; If the seller use time period and the abnormal use time period have high coincidence degree, or the seller use time period and the comparison use time period have high coincidence degree, it is determined that the seller device IP is abnormal, and it is determined that the target seller risk is high; If the seller's use period has low coincidence with the abnormal use period, and the seller's use period has low coincidence with the contrast use period, it is determined that the seller's device IP is normal, and it is determined that the target seller has low risk. 2.The method of claim 1, wherein: The evaluation strategy specifically comprises: The determination results of the target seller and the target commodity are obtained respectively; If it is determined that the target commodity has low risk, a payment interface is entered; If it is determined that the target seller has low risk, a payment interface is entered; If it is determined that the target commodity has high risk, the buyer is prompted that the commodity purchased by the buyer has risk, and the buyer is suggested to cancel the order; If it is determined that the target seller has high risk, the buyer is prompted that the seller corresponding to the commodity purchased by the buyer has risk, and the buyer is suggested to cancel the order; If the buyer chooses to continue to pay, the target seller's account is locked after the buyer pays, and the fund is unlocked after the buyer confirms that the consumption is correct; If the buyer feeds back that the consumption has problems and provides proof within the time of locking the fund, the fund is directly transferred back to the payment account of the buyer.

Images