Competitive product identification method and system based on search result ranking

Abstract

The invention discloses a competitive product identification method and system based on search result ranking. The method comprises the steps of obtaining target search data related to a target commodity; preprocessing the obtained target search data, and extracting a two-dimensional space matrix corresponding to the ranking of each to-be-identified commodity under the target hotspot keyword; fusing the search volume corresponding to each element, and calculating to obtain a final total target ranking matrix of each to-be-identified commodity combined with the search volume factor; calculating the similarity between each to-be-identified commodity and the target commodity based on Euclidean similarity, and summarizing to obtain all similarity vectors related to the similarity of the target commodity; the above steps are repeated according to a preset time period, weighted addition and sorting are carried out on the updated similarity vectors in the preset time period, commodity screening and matching are carried out on a final similar commodity list to obtain a commodity with the highest correlation as a competitive product of the target commodity, and the working efficiency and accuracy of the competitive product identification method are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of commodity information search and competitive product analysis, and in particular to a competitive product identification method and system based on search result ranking. Background Art

[0002] With the development of e-commerce and online marketing, it is crucial for companies to understand the competition of goods in the market for formulating market strategies. Traditional competitive product analysis methods are usually based on sales data, user reviews and other information. The disadvantages of this method are: 1. Delay: Sales data and user reviews are often collected after a period of time after the product is launched, resulting in the company's inability to understand market changes in real time; due to the lag of data, companies may miss the key opportunity to adjust market strategies in a timely manner and cannot respond quickly to market changes. 2. Inaccuracy: Sales data may be affected by promotional activities, seasonal changes and other factors, resulting in data fluctuations and unable to accurately reflect the product's true market performance; at the same time, user reviews may be affected by extreme emotions and personal subjective preferences, and there are deviations, which are not enough to fully reflect the product's competitive position in the market. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for identifying competing products based on search result ranking, thereby improving the working efficiency and accuracy of the competing product identification method.

[0004] The present invention provides a method for identifying competing products based on search result ranking, comprising:

[0005] Acquire target search data related to the target product, wherein the target search data at least includes a list of hot keywords related to the target product, search volumes corresponding to the target hot keywords, and ranking information of each searched product to be identified in the corresponding search results;

[0006] Preprocessing the acquired target search data, extracting the preprocessed target search data, generating a final two-dimensional ranking table of products to be identified, and extracting a two-dimensional space matrix corresponding to the ranking of each product to be identified under the target hot keyword from the final two-dimensional ranking table of products to be identified;

[0007] According to each element in the two-dimensional space matrix and the search volume corresponding to each element, the total target ranking matrix of each product to be identified combined with the search volume factor is calculated;

[0008] Obtain each product to be identified and the ranking vector corresponding to each product to be identified according to the target ranking matrix, calculate the similarity between each product to be identified and the target product based on Euclidean similarity, and summarize all similarity vectors related to the similarity of the target product;

[0009] Repeat the above steps according to a preset time period to obtain an updated similarity vector every day, perform weighted addition and sorting on the updated similarity vectors within the preset time period, obtain a final similar product list ranked by similarity to the target product, screen the products in the final similar product list and match the products with the highest correlation as competitors of the target product.

[0010] Preferably, the preprocessing of the acquired target search data includes:

[0011] Based on the e-commerce platform, a list of hot keywords related to the target product is obtained. K=[K 1 , K 2 , K a , ..., K m ] and the search volume of the corresponding hot keywords Vol = [Vol 1 , Vol. 2 , Vol. 3 ,..., Vol. m ];

[0012] The first keyword K1 in the hot keyword list K is used to search for corresponding products to obtain first search result information, wherein the first search result information includes a product list P searched according to the first keyword. 1 =[P 11 , P 12 , P 13 , ..., P 1n ] and the row vector Rank of the corresponding product in the search results 1 = [1, 2, 3, ..., n];

[0013] The corresponding product search is performed using the second keyword K2 in the hot keyword list K to obtain second search result information, wherein the second search result information includes a product list P searched according to the second keyword. 2 =[P 21 , P 22 , P 23 , ..., P 2n ] and the corresponding product's secondary product ranking row vector Rank in the search results 2 = [1, 2, 3, ..., n];

[0014] Traverse all hot keywords in the hot keyword list K, and perform a deduplication operation on the same products in all product lists after the search, and obtain a final product list P = [P 1 , P 2 , P 3 , ..., P V] and the final product ranking two-dimensional table generated after deduplication;

[0015] Among them, P 11 ,P 12 ,P 13 ,…,P 1n All are products searched based on the first keyword K1, P 21 ,P 22 ,P 23 ,…,P 2n are all products searched based on the second keyword K2, and v is the number of all non-duplicate products crawled by all hot keywords.

[0016] Preferably, according to each element in the two-dimensional space matrix and integrating the search volume corresponding to each element, the total target ranking matrix of each commodity to be identified combined with the search volume factor is calculated to include:

[0017] Extract the intermediate data in the final product ranking two-dimensional table to obtain a two-dimensional spatial ranking matrix

[0018]

[0019] Introduce the search volume vector Vol corresponding to the hot keywords, perform element-by-element multiplication of the search volume vector and the two-dimensional ranking matrix to obtain the initial similarity array

[0020]

[0021] Among them, Rank vm It is product P v In the hot keyword K m The ranking below, v is the number of all non-duplicate products crawled by all hot keywords.

[0022] Preferably, each commodity to be identified and the ranking vector corresponding to each commodity to be identified are obtained according to the target ranking matrix, and the similarity between each commodity to be identified and the target commodity is calculated based on the Euclidean similarity, and all similarity vectors related to the similarity of the target commodity are summarized, including:

[0023] Product P 1 As the target product, calculate the product P to be identified respectively 2 , P 3 ,…,P v The corresponding ranking vector S 2 , S 3 ,…,S v The target product P 1 The corresponding ranking vector S 1 The Euclidean distance of

[0024] For each k∈{2,3,…,v}, calculate the ranking vector S 1 and S k The Euclidean distance between is as follows:

[0025]

[0026] Among them, d(S 1 , S k ) is calculated as follows:

[0027]

[0028] The final similarity vector CS is obtained by summarizing the calculations:

[0029]

[0030] Among them, S 1 , S 2 , S 3 ,…,S v is the product P in the initial similarity array S 1 , P 2 , P 3 ,…,P v The corresponding ranking vector of .

[0031] Preferably, weighted addition and sorting of similarity vectors updated within a preset time period are performed to obtain a final similar product list ranked by similarity to the target product, including:

[0032] Within the preset time period p days, the final similarity vector of the product to be identified is updated every day and recorded as CS a , CS b , CS c ,…,CS p ;

[0033] If the data of the product to be identified is not found based on the target hot keywords on the day, the average or median value is used to supplement it. If the product to be identified is updated with a new search volume, the final similarity vector is adjusted;

[0034] The final similarity vectors obtained every day within p days are weighted and added, and the weight is the average weight:

[0035]

[0036] Select the product to be identified corresponding to the row where the final similarity vector value is greater than the preset similarity threshold, and output it as the competitor of the target product;

[0037] Among them, CS a , CS b , CSc ,…,CS p They are the final similarity vectors of all products to be identified obtained on the first day, the second day, ... the pth day.

[0038] Preferably, the target search data also includes a fluctuation rate of each commodity to be identified.

[0039] The present invention also provides a competitive product identification system based on search result ranking, comprising:

[0040] A data acquisition module is used to acquire target search data related to the target product, wherein the target search data at least includes a list of hot keywords related to the target product, search volumes corresponding to the target hot keywords, and ranking information of each searched product to be identified in the corresponding search results;

[0041] A preprocessing module is used to preprocess the acquired target search data, extract the preprocessed target search data, generate a final two-dimensional ranking table of products to be identified, and extract a two-dimensional space matrix corresponding to the ranking of each product to be identified under the target hot keyword from the final two-dimensional ranking table of products to be identified;

[0042] The factor fusion module calculates the total target ranking matrix of the final products to be identified by combining the search volume factors according to each element in the two-dimensional space matrix and the search volume corresponding to each element;

[0043] A similarity calculation module is used to obtain each product to be identified and the ranking vector corresponding to each product to be identified according to the target ranking matrix, calculate the similarity between each product to be identified and the target product based on Euclidean similarity, and summarize all similarity vectors related to the similarity of the target product;

[0044] The screening output module is used to repeat the calculation according to a preset time period, obtain an updated similarity vector every day, perform weighted addition and sorting on the updated similarity vectors within the preset time period, obtain a final similar product list ranked by similarity to the target product, screen the products in the final similar product list and match the products with the highest correlation as competitors of the target product.

[0045] Preferably, the screening output module includes:

[0046] The update calculation unit is used to update the final similarity vector of the product to be identified every day within a preset time period p days, denoted as CS a , CS b , CS c ,…,CS p ;

[0047] An adjustment unit, used to supplement the data of the commodity to be identified by using the average value or median value if the data is not found according to the target hot keywords on the day, and to adjust the final similarity vector if the commodity to be identified is updated to a new search volume;

[0048] The weighting unit is used to perform weighted addition on the final similarity vectors obtained every day within p days, and the weight adopts the average weight:

[0049]

[0050] A screening unit, used to select the to-be-identified products corresponding to the row where the final similarity vector value is greater than a preset similarity threshold, and output them as competitors of the target product;

[0051] Among them, CS a , CS b , CS c ,…,CS p They are the final similarity vectors of all products to be identified obtained on the first day, the second day, ... the pth day.

[0052] The present invention also provides an electronic device, comprising:

[0053] A memory, the memory being used to store a processing program;

[0054] A processor, wherein when executing the processing program, the processor implements the method for identifying competing products based on search result ranking as described in an embodiment of the present invention.

[0055] The present invention also provides a readable storage medium, on which a processing program is stored. When the processing program is executed by a processor, the method for identifying competing products based on search result ranking as described in an embodiment of the present invention is implemented.

[0056] With respect to the prior art, the present invention has the following beneficial effects:

[0057] The competitive product identification method based on search result ranking provided by the present invention can effectively collect and organize a large amount of data related to the target product by obtaining a hot keyword list, search volume and search result ranking information related to the target product and preprocessing these data, thereby ensuring the accuracy and reliability of subsequent analysis; by incorporating the search volume factor into the ranking information of each product to be identified, a total target ranking matrix combined with the search volume factor is calculated, which can more accurately reflect the influence of different search keywords on the ranking of the product to be identified, thereby improving the practicality and accuracy of the ranking data; according to the target ranking matrix, the ranking vector of each product to be identified is obtained, and the similarity between each product to be identified and the target product is calculated using Euclidean similarity, which can effectively quantify the similarity between different products and provide reliable data support for further competitive product analysis; according to the target ranking matrix, the ranking vector of each product to be identified is obtained, and the similarity between each product to be identified and the target product is calculated using Euclidean similarity, thereby effectively quantifying the similarity between different products and providing .... The calculation is repeated according to the preset time period, the similarity vector is updated every day, and the final list of similar products is obtained through weighted addition and sorting. The impact of time factors on the competitive relationship of products is taken into account, making the identification of competitive products more dynamic and accurate. The entire competitive product identification system is divided into modules such as data acquisition, preprocessing, factor fusion, similarity calculation and screening output. This modular design makes the system structure clear, easy to maintain and upgrade, and also improves the operating efficiency and stability of the system. The target search data also includes the volatility of each product to be identified, which can provide more information about market dynamics and help to more accurately evaluate the market performance and competitiveness of the product. By screening and matching the products in the final similar product list, the competitive products with the highest correlation with the target product can be accurately and quickly identified, which has important guiding significance for product strategy adjustment, market positioning and competition analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 Schematic diagram of the steps of a method for identifying competing products based on search result ranking in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] Embodiment 1

[0061] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0062] It should be noted that the concepts such as "once" and "second" mentioned in the present application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0063] It should be noted that the modifications of "one" and "plurality" mentioned in the disclosure of this application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0064] like Figure 1 As shown, an embodiment of the present invention provides a method for identifying competing products based on search result ranking, including:

[0065] S1: Acquire target search data related to the target product, wherein the target search data at least includes a list of hot keywords related to the target product, search volumes corresponding to the target hot keywords, and ranking information of each searched product to be identified in the corresponding search results;

[0066] S2: preprocessing the acquired target search data, extracting the preprocessed target search data, generating a final two-dimensional ranking table of products to be identified, and extracting a two-dimensional space matrix corresponding to the ranking of each product to be identified under the target hot keyword from the final two-dimensional ranking table of products to be identified;

[0067] S3: According to each element in the two-dimensional space matrix and the search volume corresponding to each element, the total target ranking matrix of each product to be identified combined with the search volume factor is calculated;

[0068] S4: Obtain each product to be identified and the ranking vector corresponding to each product to be identified according to the target ranking matrix, calculate the similarity between each product to be identified and the target product based on Euclidean similarity, and summarize all similarity vectors related to the similarity of the target product;

[0069] S5: Repeat the above steps according to a preset time period, obtain an updated similarity vector every day, perform weighted addition and sorting on the updated similarity vectors within the preset time period, obtain a final similar product list ranked by similarity to the target product, screen the products in the final similar product list and match the products with the highest correlation as competitors of the target product.

[0070] Those skilled in the art can understand that the above embodiment first collects a list of keywords related to the target product, the search volume of these keywords and the search ranking information of each product to be identified under these keywords, processes the collected data, generates a two-dimensional table, lists the ranking of each product to be identified under each keyword, and further extracts this ranking information into a two-dimensional space matrix, and uses the two-dimensional space matrix and the search volume of each keyword to calculate a comprehensive ranking matrix, which reflects the overall ranking of each product after considering the influence of the search volume, obtains the ranking vector of each product according to the comprehensive ranking matrix, and then uses Euclidean similarity to calculate the similarity between these vectors and the target product, and summarizes the similarity vector, repeats the above process regularly such as every day, updates the similarity vector, and determines the final list of competing products by weighted summation and sorting, from which the products with the strongest competitive relationship with the target product are screened out, which is conducive to accurately identifying competing products, reducing latency, and improving the accuracy of competing product identification. For example, the above process is as follows: assuming the target product A, there are many unidentified products B and C for the search of target product A, and 10 hot keywords are entered. The distance between A and B has been searched by 10,000 people, and the distance between A and C has been searched by only 1 person, then B is considered to have a high match with her.

[0071] In this embodiment, the preprocessing of the acquired target search data in step S2 includes:

[0072] Based on the e-commerce platform, a list of hot keywords related to the target product is obtained. K=[K 1 , K 2 , K 3 , ..., K m ] and the search volume of the corresponding hot keywords Vol = [Vol 1 ,Vol 2 ,Vol 3 ,...,Vol m ]; After searching for products for each keyword in the keyword list K, a product information list will appear, and the specific products and the ranking of each product in the search results can be extracted. By using actual search data from the e-commerce platform, the real-time and accuracy of the data is ensured, making the competitive product analysis results more reliable.

[0073] The first keyword K1 in the hot keyword list K is used to search for corresponding products to obtain first search result information, wherein the first search result information includes a product list P searched according to the first keyword. 1 =[P 11 ,P 12 ,P 13 , ..., P 1n ] and the row vector Rank of the corresponding product in the search results 1 =[1,2,3,...,n];

[0074] The corresponding product search is performed using the second keyword K2 in the hot keyword list K to obtain second search result information, wherein the second search result information includes a product list P searched according to the second keyword. 2 =[P 21 ,P 22 ,P 23 ,...,P 2n ] and the corresponding product's secondary product ranking row vector Rank in the search results 2 =[1,2,3,...,n];

[0075] Traverse all hot keywords in the hot keyword list K, and perform a deduplication operation on the same products in all product lists after the search, and obtain a final product list P = [P 1 ,P 2 ,P 3 ,...,P v ] and the final product ranking two-dimensional table generated after deduplication is as follows: Table 1:

[0076]

[0077] Table 1 Two-dimensional table of product rankings

[0078] Among them, P 11 ,P 12 ,P 13 ,…,P 1n All are products searched based on the first keyword K1, P 21 ,P 22 ,P 23 ,…,P 2n They are all products searched based on the second keyword K2, v is the number of all non-duplicate products crawled by all hot keywords, and m is the product serial number.

[0079] The above process ensures that each product is only counted once, thereby obtaining the final product list and product ranking two-dimensional table. By removing duplicates from the search results, it is ensured that data analysis will not be biased due to duplicate products, thereby improving the data processing quality and analysis accuracy; the automated search and information extraction process reduces manual operations and improves the efficiency of data processing.

[0080] In this embodiment, in order to accurately identify competing products, the influencing factors related to the target product are comprehensively considered. In step S3, the total target ranking matrix of the final products to be identified combined with the search volume factor is calculated based on each element in the two-dimensional space matrix and the search volume corresponding to each element, including:

[0081] Extract the intermediate data in the final product ranking two-dimensional table to obtain a two-dimensional spatial ranking matrix

[0082]

[0083] Introduce the search volume vector Vol corresponding to the hot keywords, perform element-by-element multiplication of the search volume vector and the two-dimensional ranking matrix to obtain the initial similarity array

[0084]

[0085] Among them, Rank vm It is product P v In the hot keyword K m The ranking under , v is the number of all non-repeated products crawled by all hot keywords, which are directly integrated into the ranking calculation of each product under the keyword, thereby generating an initial similarity array. This array comprehensively considers the ranking of the product and the popularity of the keyword, that is, the search volume. By incorporating the search volume as a weight into the ranking calculation, the popularity of the product among consumers can be more accurately evaluated, not just the ranking position; the use of search volume data reduces the uncertainty and limitations of relying solely on ranking data, and improves the comprehensiveness and utilization of data analysis; through ranking, it is ensured that the competitive product analysis process takes into account both the market performance of the product and the query preferences of consumers through search volume, making the analysis results more comprehensive and practical.

[0086] In this embodiment, in step S4, each commodity to be identified and the ranking vector corresponding to each commodity to be identified are obtained according to the target ranking matrix, and the similarity between each commodity to be identified and the target commodity is calculated based on the Euclidean similarity, and all similarity vectors related to the similarity of the target commodity are summarized, including:

[0087] Product P 1 As the target product, calculate the product P to be identified respectively 2 , P 3 ,…,Pv The corresponding ranking vector S 2 , S 3 ,…,S v The target product P 1 The corresponding ranking vector S 1 The Euclidean distance of

[0088] For each k∈{2,3,…,v}, calculate the ranking vector S 1 and S k The Euclidean distance between is as follows:

[0089]

[0090] Among them, d(S 1 , S k ) is calculated as follows:

[0091]

[0092] The final similarity vector CS is obtained by summarizing the calculations:

[0093]

[0094] Among them, S 1 , S 2 , S 3 ,…,S v is the product P in the initial similarity array S 1 , P 2 , P 3 ,…,P v The corresponding ranking vector of .

[0095] In this embodiment, in step S5, weighted addition and sorting are performed on the similarity vectors updated within the preset time period to obtain a final similar product list ranked by similarity to the target product, including:

[0096] Within the preset time period p days, the final similarity vector of the product to be identified is updated every day and recorded as CS a , CS b , CS c ,…,CS p ;

[0097] If the data of the product to be identified is not found based on the target hot keywords on the day, the average or median value is used to supplement it. If the product to be identified is updated with a new search volume, the final similarity vector is adjusted. For example, if there were only 10 products to be identified before and there is a new product to be identified today, just add one row.

[0098] The final similarity vectors obtained every day within p days are weighted and added, and the weight is the average weight:

[0099]

[0100] Select the product to be identified corresponding to the row where the final similarity vector value is greater than the preset similarity threshold, and output it as the competitor of the target product;

[0101] Among them, CS a , CS b , CS c ,…,CS p They are the final similarity vectors of all products to be identified obtained on the first day, the second day, ... the pth day.

[0102] In another embodiment, the target search data also includes the volatility of each commodity to be identified. During implementation, the target is configured: based on commodity A, by searching the name, ranking in the search results, etc., to find competing products B, C, D... The conditions that can be obtained are: 1. Search keyword W, 2. Number of searches for keyword W, 3. Search time T, 4. A's ranking in the search results R_a, 5. B's ranking in the search results R_b..., 6. Set volatility σ. At time T, according to the search results, based on the ranking R_a of commodity A and the volatility σ, calculate the similar ranking interval [R_a-2σ, R_a+2σ] of commodity A, identify the commodities entering the ranking interval of the target commodity, take out these commodities and calculate their number of occurrences, and obtain a two-dimensional array:

[0103] At time T+1, repeat the above steps to obtain the two-dimensional array M2... Add the obtained two-dimensional data to get the summarized result Sort the array in descending order according to the second column, and select the top five products as competitors of product A.

[0104] Embodiment 2

[0105] Based on the same inventive concept, the present invention provides 7. A competitive product identification system based on search result ranking, characterized in that it includes:

[0106] A data acquisition module is used to acquire target search data related to the target product, wherein the target search data at least includes a list of hot keywords related to the target product, search volumes corresponding to the target hot keywords, and ranking information of each searched product to be identified in the corresponding search results;

[0107] A preprocessing module is used to preprocess the acquired target search data, extract the preprocessed target search data, generate a final two-dimensional ranking table of products to be identified, and extract a two-dimensional space matrix corresponding to the ranking of each product to be identified under the target hot keyword from the final two-dimensional ranking table of products to be identified;

[0108] The factor fusion module calculates the total target ranking matrix of the final products to be identified by combining the search volume factors according to each element in the two-dimensional space matrix and the search volume corresponding to each element;

[0109] A similarity calculation module is used to obtain each product to be identified and the ranking vector corresponding to each product to be identified according to the target ranking matrix, calculate the similarity between each product to be identified and the target product based on Euclidean similarity, and summarize all similarity vectors related to the similarity of the target product;

[0110] The screening output module is used to repeat the calculation according to a preset time period, obtain an updated similarity vector every day, perform weighted addition and sorting on the updated similarity vectors within the preset time period, obtain a final similar product list ranked by similarity to the target product, screen the products in the final similar product list and match the products with the highest correlation as competitors of the target product.

[0111] In this embodiment, the screening output module includes:

[0112] The update calculation unit is used to update the final similarity vector of the product to be identified every day within a preset time period p days, denoted as CS a , CS b , CS c ,…,CS p ;

[0113] An adjustment unit, used to supplement the data of the commodity to be identified by using the average value or median value if the data is not found according to the target hot keywords on the day, and to adjust the final similarity vector if the commodity to be identified is updated to a new search volume;

[0114] The weighting unit is used to perform weighted addition on the final similarity vectors obtained every day within p days, and the weight adopts the average weight:

[0115]

[0116] A screening unit, used to select the to-be-identified products corresponding to the row where the final similarity vector value is greater than a preset similarity threshold, and output them as competitors of the target product;

[0117] Among them, CS a , CS b , CSc ,…,CS p They are the final similarity vectors of all products to be identified obtained on the first day, the second day, ... the pth day.

[0118] The principles, contents and implementation methods of the above modules and units are all as described in Example 1 and will not be repeated here.

[0119] Embodiment 3

[0120] In some embodiments of the present application, an electronic device is also provided. The electronic device includes a memory and a processor, wherein the memory is used to store a processing program, and the processor executes the processing program according to instructions. When the processor executes the processing program, the method for identifying competitive products based on search result ranking in the above-mentioned embodiment is implemented.

[0121] In some embodiments of the present application, a readable storage medium is further provided, which may be a non-volatile readable storage medium or a volatile readable storage medium. The readable storage medium stores instructions, which, when executed on a computer, cause an electronic device including the readable storage medium to execute the aforementioned method for identifying competing products based on search result ranking.

[0122] It is understandable that, for the aforementioned competitive product identification method based on search result ranking, if it is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0123] Computer readable storage media may include data signals propagated in baseband or as part of a carrier wave, wherein readable program codes are carried. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program codes contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0124] The program code involved in executing the technical solution disclosed in the present application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and also conventional procedural programming languages ​​such as C language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on the remote computing device, or completely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying competing products based on search result ranking, characterized in that: include: Acquire target search data related to the target product, wherein the target search data at least includes a list of hot keywords related to the target product, search volumes corresponding to the target hot keywords, and ranking information of each searched product to be identified in the corresponding search results; Preprocessing the acquired target search data, extracting the preprocessed target search data, generating a final two-dimensional ranking table of products to be identified, and extracting a two-dimensional space matrix corresponding to the ranking of each product to be identified under the target hot keyword from the final two-dimensional ranking table of products to be identified; According to each element in the two-dimensional space matrix and the search volume corresponding to each element, the total target ranking matrix of each product to be identified combined with the search volume factor is calculated; Obtain each product to be identified and the ranking vector corresponding to each product to be identified according to the target ranking matrix, calculate the similarity between each product to be identified and the target product based on Euclidean similarity, and summarize all similarity vectors related to the similarity of the target product; Repeat the above steps according to a preset time period to obtain an updated similarity vector every day, perform weighted addition and sorting on the updated similarity vectors within the preset time period, obtain a final similar product list ranked by similarity to the target product, screen the products in the final similar product list and match the products with the highest correlation as competitors of the target product.

2. The method for identifying competing products based on search result ranking according to claim 1, characterized in that: The preprocessing of the acquired target search data includes: Based on the e-commerce platform, a list of hot keywords K=[K1, K2, K3, ..., K m ] and the search volume of the corresponding hot keywords Vol = [Vol1, Vol2, Vol3, ..., Vol m ]; The first keyword K1 in the hot keyword list K is used to search for corresponding products to obtain first search result information, wherein the first search result information includes a product list P1=[P 11 , P 12 , P 13 , ..., P 1n ] and the product ranking row vector Rank1 = [1, 2, 3, ..., n] of the corresponding product in the search results; The corresponding product search is performed using the second keyword K2 in the hot keyword list K to obtain second search result information, wherein the second search result information includes a product list P2=[P 21 , P 22 , P 23 ,…,P 2n ] and the secondary product ranking row vector Rank2 = [1, 2, 3, ..., n] of the corresponding product in the search results; Traverse all hot keywords in the hot keyword list K, and perform a deduplication operation on the same products in all product lists after the search, and obtain a final product list P = [P1, P2, P3, ..., P V ] and the final product ranking two-dimensional table generated after deduplication; Among them, P 11 ,P 12 ,P 13 ,…,P 1n All are products searched based on the first keyword K1, P 21 ,P 22 ,P 23 ,…,P 2n are all products searched based on the second keyword K2, and v is the number of all non-duplicate products crawled by all hot keywords.

3. The method for identifying competing products based on search result ranking according to claim 1, characterized in that: According to each element in the two-dimensional space matrix and the search volume corresponding to each element, the total target ranking matrix of each product to be identified combined with the search volume factor is calculated, including: Extract the intermediate data in the final product ranking two-dimensional table to obtain a two-dimensional spatial ranking matrix Introduce the search volume vector Vol corresponding to the hot keywords, perform element-by-element multiplication of the search volume vector and the two-dimensional ranking matrix to obtain the initial similarity array Among them, Rank vm It is product P v In the hot keyword K m The ranking below, v is the number of all non-duplicate products crawled by all hot keywords.

4. The method for identifying competing products based on search result ranking according to claim 1, characterized in that: According to the target ranking matrix, each product to be identified and the ranking vector corresponding to each product to be identified are obtained, and the similarity between each product to be identified and the target product is calculated based on the Euclidean similarity. All similarity vectors related to the similarity of the target product are summarized and include: Take product P1 as the target product and calculate the products to be identified P2, P3, ..., P v The corresponding ranking vectors S2, S3, ..., S v The Euclidean distance of the ranking vector S1 corresponding to the target product P1; For each k∈{2,3,…,v}, calculate the ranking vectors S1 and S k The Euclidean distance between is as follows: Among them, d(S1,S k ) is calculated as follows: The final similarity vector CS is obtained by summarizing the calculations: Among them, S1, S2, S3, ..., S v are the products P1, P2, P3, ..., P in the initial similarity array S v The corresponding ranking vector of .

5. The method for identifying competing products based on search result ranking according to claim 1, characterized in that: The similarity vectors updated within a preset time period are weighted added and sorted to obtain a final similar product list ranked by similarity to the target product, including: Within the preset time period p days, the final similarity vector of the product to be identified is updated every day and recorded as CS a , CS b , CS c ,…,CS p ; If the data of the product to be identified is not found based on the target hot keywords on the day, the average or median value is used to supplement it. If the product to be identified is updated with a new search volume, the final similarity vector is adjusted; The final similarity vectors obtained every day within p days are weighted and added, and the weight is the average weight: Select the product to be identified corresponding to the row where the final similarity vector value is greater than the preset similarity threshold, and output it as the competitor of the target product; Among them, CS a , CS b , CS c ,…,CS p They are the final similarity vectors of all products to be identified obtained on the first day, the second day, ... the pth day.

6. The method for identifying competing products based on search result ranking according to claim 1, characterized in that: The target search data also includes a fluctuation rate of each commodity to be identified.

7. A competitive product identification system based on search result ranking, characterized in that: include: A data acquisition module is used to acquire target search data related to the target product, wherein the target search data at least includes a list of hot keywords related to the target product, search volumes corresponding to the target hot keywords, and ranking information of each searched product to be identified in the corresponding search results; A preprocessing module is used to preprocess the acquired target search data, extract the preprocessed target search data, generate a final two-dimensional ranking table of products to be identified, and extract a two-dimensional space matrix corresponding to the ranking of each product to be identified under the target hot keyword from the final two-dimensional ranking table of products to be identified; The factor fusion module calculates the total target ranking matrix of the final products to be identified by combining the search volume factors according to each element in the two-dimensional space matrix and the search volume corresponding to each element; A similarity calculation module is used to obtain each product to be identified and the ranking vector corresponding to each product to be identified according to the target ranking matrix, calculate the similarity between each product to be identified and the target product based on Euclidean similarity, and summarize all similarity vectors related to the similarity of the target product; The screening output module is used to repeat the calculation according to a preset time period, obtain an updated similarity vector every day, perform weighted addition and sorting on the updated similarity vectors within the preset time period, obtain a final similar product list ranked by similarity to the target product, screen the products in the final similar product list and match the products with the highest correlation as competitors of the target product.

8. The competitive product identification system based on search result ranking according to claim 7, characterized in that: The filter output modules include: The update calculation unit is used to update the final similarity vector of the product to be identified every day within a preset time period p days, denoted as CS a , CS b , CS c ,…,CS p ; An adjustment unit, used to supplement the data of the commodity to be identified by using the average value or median value if the data is not found according to the target hot keywords on the day, and to adjust the final similarity vector if the commodity to be identified is updated to a new search volume; The weighting unit is used to perform weighted addition on the final similarity vectors obtained every day within p days, and the weight adopts the average weight: A screening unit, used to select the to-be-identified products corresponding to the row where the final similarity vector value is greater than a preset similarity threshold, and output them as competitors of the target product; Among them, CS a , CS b , CS c ,…,CS p They are the final similarity vectors of all products to be identified obtained on the first day, the second day, ... the pth day.

9. An electronic device, characterized in that: include: A memory, the memory being used to store a processing program; A processor, wherein when executing the processing program, the processor implements the method for identifying competing products based on search result ranking as described in any one of claims 1 to 6.

10. A readable storage medium, characterized in that: The readable storage medium stores a processing program, and when the processing program is executed by a processor, the method for identifying competing products based on search result ranking as described in any one of claims 1 to 6 is implemented.

Images