Analytical method, device, electronic device and readable storage medium for a product

Abstract

The application provides a product analysis method and device, electronic equipment and a readable storage medium. The product analysis method comprises: obtaining consultation record information of a target user in a preset time period; determining at least one second product related to a first product according to the consultation record information; determining a first consultation probability of the first product and a second consultation probability of the second product according to the consultation record information and N preset consultation information, wherein N is an integer greater than 0; and determining difference information and homogeneous information of the first product and the second product from the N preset consultation information according to the first consultation probability and the second consultation probability. The technical scheme provided by the application realizes the mining of the relationship between products by statistically analyzing the consultation session data of the same user for the same type of product in the same time period, obtaining the intention of the same user to consult a specific product and potential competitors, and determining the difference points and homogeneous points of the specific product and potential competitors.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically, to a product analysis method, a product analysis device, an electronic device, and a readable storage medium. Background Technology

[0002] In related technologies, big data analysis of all user behavior information and product reviews is used to identify and compare competitors, thereby determining user attitudes towards the product. However, this method of comparing products with competitors based on data from all users is somewhat one-sided, resulting in low accuracy in the comparison. Summary of the Invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art or related art.

[0004] Therefore, a first aspect of the present invention provides a method for analyzing a product.

[0005] A second aspect of the present invention also provides an analytical apparatus for a product.

[0006] A third aspect of the present invention also provides an electronic device.

[0007] A fourth aspect of the present invention also provides an electronic device.

[0008] A fifth aspect of the invention also provides a readable storage medium.

[0009] In view of this, a first aspect of the present invention provides a product analysis method, comprising: acquiring consultation record information of a target user within a preset time period; determining at least one second product related to a first product based on the consultation record information; determining a first consultation probability of the first product and a second consultation probability of the second product based on the consultation record information and N preset consultation information, wherein N is an integer greater than 0; and determining difference information and homogeneity information of the first product and the second product from the N preset consultation information based on the first consultation probability and the second consultation probability.

[0010] The product analysis method provided by this invention, for a specific product (i.e., the first product), obtains the target users and times that initiate inquiries about it on e-commerce platforms, and based on the time of the inquiries, obtains the inquiry record information of the target users within a preset time period (e.g., 24 hours before and after the inquiry time), and identifies at least one potential competitor (i.e., the second product) that is a product of the same category as the specific product from the inquiry record information.

[0011] Furthermore, based on N consultation intentions (i.e., preset consultation information), statistical analysis is performed on the consultation record information to obtain the consultation intentions of target users regarding specific products. Using these consultation intentions, the degree of preference of target users for consultation intentions regarding specific products (i.e., the first consultation probability) is analyzed. Additionally, the degree of preference of target users for consultation intentions regarding potential competitors is analyzed using these consultation intentions.

[0012] Furthermore, the analysis examines the target users' preferences for specific products and their preferences for potential competitors. From N inquiry intentions, the analysis identifies the unique features of the product that cannot be replaced by potential competitors (i.e., differentiating information) and the similarities that the product can be replaced by potential competitors (i.e., homogeneous information) for the target users.

[0013] This application statistically analyzes consultation session data of the same user regarding the same type of product within the same time period to obtain the user's consultation intent for specific products and potential competitors. Using this consultation intent, it determines the user's preference for specific products and potential competitors, further analyzes the user's emotional trajectory towards these products, calculates the competitive relationship between products, and identifies the differences and similarities between specific products and potential competitors, thus achieving the discovery of competitive relationships between products. Compared to existing technologies that rely on acquiring large amounts of data from different users' access and search reviews to identify competitors, this application constructs a product comparison combination consistent with the same user's decision-making behavior, achieving consistency in competitor comparison and effectively improving the accuracy of comparing specific products with potential competitors.

[0014] It should be noted that the N preset consultation messages are effective consultation intentions for the product summarized in advance based on the model. They are used to determine in advance the user's intention to make a certain response to the product, representing a preparatory state before taking action in the consumer behavior process. Among them, consultation intentions can be pre-sales questions, such as product introductions and preferential policies, or post-sales questions, such as logistics tracking and how to return or exchange goods.

[0015] The analytical method for the product provided by the present invention may also have the following additional technical features:

[0016] In the above technical solution, the step of determining at least one second product corresponding to the first product based on the consultation record information specifically includes: obtaining the category information of the first product; and determining at least one second product related to the first product in the consultation record information based on the category information.

[0017] In this technical solution, based on the category information of a specific product, the product combination within the same product category that is closest to the category information of the specific product is obtained. Then, at least one product from the product combination appearing in the consultation record information is identified as a potential competitor. By statistically analyzing the consultation record information of the same user regarding the same product category within a preset time period, potential competitors related to a specific product are identified, improving the accuracy of potential competitor selection, reducing the impact of irrelevant products on the analysis of a specific product, and thus improving the accuracy of product analysis.

[0018] Specifically, product category information includes brand, category, price, weight, color, size, material, place of origin, packaging, and logistics, etc., which are not limited in this application.

[0019] In any of the above technical solutions, the step of determining the first consultation probability of the first product based on consultation record information and N preset consultation information specifically includes: obtaining the first consultation record information related to the first product from the consultation record information; determining the first frequency of occurrence of each preset consultation information from the first consultation record information; and calculating the first consultation probability based on the first frequency and a first formula.

[0020] The first formula is:

[0021]

[0022] Wherein, e refers to the first product, and I refers to the first product. e For all preset consultation information appearing in the first consultation record information, the above I ek For the kth preset consultation information appearing in the first consultation record information, the above A ek Let F(I) be the probability of the k-th preset consultation information appearing in the first consultation record information. ek ) represents the frequency of the k-th preset consultation message. The sum of the frequencies of all preset consultation information, k∈{1,2,3,…,N}.

[0023] In this technical solution, the consultation record information is statistically analyzed to obtain the consultation records of the target user for the specific product (i.e., the first consultation record information). Using this consultation record information, all consultation intentions of the target user for the specific product are determined, the frequency of each consultation intention is calculated, and the sum of the frequencies of all consultation intentions is calculated.

[0024] Furthermore, the probability of each consultation intent is calculated using the formula, which represents the target user's preference level for each consultation intent related to a specific product. By calculating the target user's preference level for a specific product, the user's attention to characteristic products is determined, enabling analysis of the target user's emotional trajectory towards specific products and driving product optimization and upgrades.

[0025] In any of the above technical solutions, the step of determining the second consultation probability of the second product based on consultation record information and N preset consultation information specifically includes: obtaining second consultation record information related to the second product from the consultation record information; determining the second frequency of occurrence of each preset consultation information from the second consultation record information; and calculating the second consultation probability based on the second frequency and the second formula.

[0026] The second formula is:

[0027]

[0028] Wherein, j is the second product, and i is the third product. j For all preset consultation information appearing in the second consultation record information, the above i jk For the kth preset consultation information appearing in the second consultation record information, the above a jk Let F(i) be the probability of the k-th preset consultation information appearing in the second consultation record information. jk ) represents the frequency of the k-th preset consultation message. The sum of the frequencies of all preset consultation information, k∈{1,2,3,…,N}.

[0029] In this technical solution, the consultation record information is statistically analyzed to obtain the consultation records of the target user regarding potential competitors (i.e., the second consultation record information). Using this consultation record, all consultation intentions of the target user regarding potential competitors are determined, the frequency of each consultation intention is calculated, and the sum of the frequencies of all consultation intentions is calculated.

[0030] Furthermore, the probability of each consultation intent occurring is calculated using the formula, which represents the target user's preference for consultation intent related to potential competitors. By calculating the target user's preference for consultation intent related to potential competitors, the level of attention the target user pays to potential competitors is determined, thereby enabling the analysis of the target user's emotional trajectory towards potential competitors.

[0031] In any of the above technical solutions, the step of determining the difference information and homogeneity information of the first product and the second product from N preset consultation information based on the first consultation probability and the second consultation probability specifically includes: determining the difference value of the first product and the second product based on the first consultation probability and the second consultation probability; and determining the difference information and homogeneity information of the first product and the second product from N preset consultation information based on the difference value, the first consultation probability and the second consultation probability.

[0032] This technical solution determines the difference between the specific product and potential competitors based on the target user's preference for consulting about a specific product and their preference for consulting about potential competitors. Using these difference values, the target user's preference for the specific product, and their preference for potential competitors, the solution analyzes the points where the specific product cannot be replaced by potential competitors, as well as the points where the specific product can be replaced by potential competitors. Compared to existing methods that superficially analyze competitors based on the behavioral chains of all users, this application uses the same user's mindset and motivations to conduct a more precise analysis of competitors, achieving consistency in competitor comparison and improving the accuracy of obtaining differences and similarities between products.

[0033] In any of the above technical solutions, the difference value between the first product and the second product is further determined based on the first consultation probability and the second consultation probability, specifically including: the difference value is 0 based on the second consultation probability being equal to 0; the difference value is calculated according to the third formula based on the second consultation probability not being equal to 0.

[0034] The third formula is:

[0035]

[0036] Among them, the above L ejk The difference between the first product and the second product, A above. ek As the probability of the first consultation, the above a jk This represents the probability of the second consultation.

[0037] In this technical solution, statistical analysis is performed on the consultation records corresponding to potential competitors. If the frequency of a certain consultation intent is 0, it means that the target user's preference for that consultation intent regarding the potential competitor is 0, and the difference between the specific product and the potential competitor for that consultation intent is 0.

[0038] Furthermore, if the frequency of a particular consultation intent is not zero, meaning the target user's preference for that consultation intent regarding a potential competitor is not zero, the ratio of the target user's preference for that consultation intent regarding a specific product to their preference for that consultation intent regarding a potential competitor is calculated. This ratio represents the difference between the specific product and the potential competitor, thus determining the difference values ​​for all consultation intents. By calculating the difference values ​​for the same consultation intent across products, the accuracy of the comparative analysis between the specific product and potential competitors is effectively improved.

[0039] In any of the above technical solutions, the step of determining the difference information and homogeneity information of the first product and the second product from N preset consultation information based on the difference value, the first consultation probability, and the second consultation probability specifically includes: determining the first consultation information that meets the first preset condition from the N preset consultation information; using the first consultation information as the difference information of the first product and the second product, wherein the first preset condition is that the absolute value of the difference between the difference value and the first threshold is greater than the second threshold, the first consultation probability is greater than the third threshold, and the second consultation probability is greater than the third threshold; determining the second consultation information that meets the second preset condition from the N preset consultation information; using the second consultation information as the homogeneity information of the first product and the second product, wherein the second preset condition is that the absolute value of the difference between the difference value and the first threshold is less than the fourth threshold, the first consultation probability is greater than the third threshold, and the second consultation probability is greater than the third threshold; wherein 0 < fourth threshold < second threshold < 1.

[0040] In this technical solution, the difference between any consultation intention and the first threshold is calculated. If the absolute value of the difference is less than the difference threshold (i.e., the second threshold), the frequency of the consultation intention in the consultation records of the characteristic product is greater than the frequency threshold (i.e., the third threshold), and the frequency of the consultation intention in the consultation records of the potential product is greater than the frequency threshold, it indicates that for the target user, the consultation intention is a difference point that the specific product cannot be replaced by the potential product.

[0041] Furthermore, the difference between any consultation intent and the first threshold is calculated. If the absolute value of the difference is greater than the homogeneity threshold (i.e., the fourth threshold), the frequency of this consultation intent appearing in the consultation records of the featured product is greater than the frequency threshold, and the frequency of its appearance in the consultation records of the potential product is also greater than the frequency threshold. This indicates that, for the target user, this consultation intent represents a homogeneous point where the specific product can be replaced by the potential product. This application, based on the same comparison standard, determines the differences and homogeneities between products, achieving consistency in competitor comparison and improving the accuracy of selecting differences and homogeneities.

[0042] It should be noted that the first threshold is 1. The absolute value of the difference value minus 1 is used as the judgment value for difference points and homogeneity points. The closer the judgment value corresponding to any consultation intention is to 0, the more similar the user's preference for that consultation intention is for the specific product and potential competitors. When the judgment value is less than the homogeneity threshold, and the frequency of occurrence of that consultation intention is greater than the frequency threshold for both the featured product and potential competitors, the consultation intention corresponding to that judgment value is determined to be a homogeneity point between the specific product and potential competitors. The further the judgment value corresponding to any consultation intention is from 0, the greater the difference in the user's preference for that consultation intention between the featured product and potential competitors. When the judgment value is greater than the difference threshold, and the frequency of occurrence of that consultation intention is greater than the frequency threshold for both the featured product and potential competitors, the consultation intention is determined to be a difference point between the specific product and potential competitors. By limiting the size of the difference threshold and the homogeneity threshold, the consistency of the judgment logic is ensured, effectively improving the accuracy of product analysis. Specifically, 0 < homogeneity threshold < difference threshold < 1. By limiting the difference threshold and the homogeneity threshold, the consistency of the judgment logic is ensured, improving the accuracy of product analysis.

[0043] Furthermore, the values ​​of the second, third, and fourth thresholds are set according to the actual situation, such as the number of homogeneous and differential information required, and are not limited in this application.

[0044] In any of the above technical solutions, when there are multiple second products, the method further includes: determining the similarity between the first product and each second product based on the first consultation probability and the second consultation probability; and determining the second product with a similarity greater than the similarity threshold from the multiple second products.

[0045] In this technical solution, when there are multiple potential competitors, the similarity between the specific product and each potential competitor's consultation intent is determined based on the target user's preference for the specific product and their preference for the potential competitors' consultation intent. Multiple similarity values ​​are then ranked; higher similarity indicates a smaller difference in the target user's consultation intent between the two products, meaning the potential competitors and the specific product are more similar based on the target user's preferences. Conversely, lower similarity indicates a greater difference in the target user's consultation intent between the two products, meaning the potential competitors and the specific product are less similar based on the target user's preferences. By comparing the similarity of the target user's consultation intent between the specific product and the potential competitors, the difference in the target user's attention to the two products is determined, improving the accuracy of the analysis of the degree of competitiveness between products.

[0046] According to a second aspect of the present invention, a product analysis apparatus is provided, comprising: an acquisition module for acquiring consultation record information of a target user within a preset time period; a first determination module for determining at least one second product related to a first product based on the consultation record information; a second determination module for determining a first consultation probability of the first product and a second consultation probability of the second product based on the consultation record information and N preset consultation information, wherein N is an integer greater than 0; and a third determination module for determining difference information and homogeneity information between the first product and the second product from the N preset consultation information based on the first consultation probability and the second consultation probability.

[0047] In this technical solution, the product analysis device includes an acquisition module, a first determination module, a second determination module, and a third determination module.

[0048] Specifically, the acquisition module is configured to acquire the target users and times that initiate inquiries about a specific product (i.e., the first product) on the e-commerce platform, and based on the time of the inquiries, acquire the inquiry record information of the target users within a preset time period (e.g., 24 hours before and after the inquiry time).

[0049] Furthermore, the first determining module is configured to identify at least one potential competitor (i.e., a second product) that is a product of the same category as the specific product from the consultation record information.

[0050] Furthermore, the second determining module is configured to perform statistical analysis on the consultation record information based on N consultation intentions (i.e., preset consultation information), obtain the consultation intentions of target users for consulting about specific products appearing in the consultation record information, analyze the degree of preference of target users for consulting about specific products (i.e., the first consultation probability) using the consultation intentions, and analyze the degree of preference of target users for consulting about potential competitors appearing in the consultation record information (i.e., the second consultation probability) using the consultation intentions.

[0051] Furthermore, the third determination module is configured to analyze the target user's preference for a specific product and their preference for potential competitors. From N consultation intentions, it identifies the differences (i.e., difference information) that the featured product cannot be replaced by potential competitors for the target user, and the homogeneous points (i.e., homogeneous information) that the featured product can be replaced by potential competitors.

[0052] This application statistically analyzes consultation session data of the same user regarding the same type of product within the same time period to obtain the user's consultation intent for specific products and potential competitors. Using this consultation intent, it determines the user's preference for specific products and potential competitors, further analyzes the user's emotional trajectory towards these products, calculates the competitive relationship between products, and identifies the differences and similarities between specific products and potential competitors, thus achieving the discovery of competitive relationships between products. Compared to existing technologies that rely on acquiring large amounts of data from different users' access and search reviews to identify competitors, this application constructs a product comparison combination consistent with the same user's decision-making behavior, achieving consistency in competitor comparison and effectively improving the accuracy of comparing specific products with potential competitors.

[0053] According to a third aspect of the present invention, an electronic device is provided, comprising an analysis apparatus for the product proposed in the second aspect. Therefore, this electronic device possesses all the beneficial effects of the analysis apparatus for the product proposed in the second aspect, which will not be elaborated further here.

[0054] According to a fourth aspect of the present invention, an electronic device is provided, comprising: a memory storing a program or instructions; and a processor connected to the memory, configured to implement the analysis method of the product proposed in the first aspect when executing the program or instructions. Therefore, this electronic device possesses all the beneficial effects of the analysis method of the product proposed in the first aspect, which will not be elaborated further here.

[0055] According to a fifth aspect of the present invention, a readable storage medium is provided on which a program or instructions are stored, which, when executed by a processor, perform the analysis method of the product proposed in the first aspect. Therefore, this readable storage medium possesses all the beneficial effects of the analysis method of the product proposed in the first aspect, and to avoid repetition, further details are omitted.

[0056] Additional aspects and advantages of the invention will become apparent in the following description or may be learned by practice of the invention. Attached Figure Description

[0057] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0058] Figure 1 This illustration shows one of the schematic flowcharts of a product analysis method according to an embodiment of the present invention;

[0059] Figure 2 This is a second schematic diagram of a product analysis method flow according to an embodiment of the present invention;

[0060] Figure 3This is a schematic diagram of the product analysis method flow according to an embodiment of the present invention (Part 3).

[0061] Figure 4 A fourth schematic diagram of the product analysis method flow according to an embodiment of the present invention is shown;

[0062] Figure 5 The fifth illustration shows a schematic flowchart of a product analysis method according to an embodiment of the present invention;

[0063] Figure 6 A schematic diagram of the product analysis method flow according to an embodiment of the present invention is shown in Figure 6.

[0064] Figure 7 This is illustrated in diagram seven of the product analysis method flowcharts of an embodiment of the present invention;

[0065] Figure 8 This is illustrated as a schematic diagram of the product analysis method according to an embodiment of the present invention (Figure 8).

[0066] Figure 9 A schematic diagram of the product analysis method flow according to an embodiment of the present invention is shown in Figure 9.

[0067] Figure 10 A schematic diagram illustrating a product analysis method according to a specific embodiment of the present invention is shown;

[0068] Figure 11 One of the schematic block diagrams illustrating a product analysis method according to a specific embodiment of the present invention is shown;

[0069] Figure 12 The second schematic block diagram illustrates a product analysis method according to a specific embodiment of the present invention;

[0070] Figure 13 A schematic block diagram of an analysis apparatus for a product according to an embodiment of the present invention is shown;

[0071] Figure 14 A schematic block diagram of an electronic device according to the present invention is shown.

[0072] in, Figure 13 and Figure 14 The correspondence between the reference numerals and component names in the attached drawings is as follows:

[0073] 1300 Product analysis device, 1302 Acquisition module, 1304 First determination module, 1306 Second determination module, 1308 Third determination module, 1400 Electronic device, 1402 Memory, 1404 Processor. Detailed Implementation

[0074] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0075] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0076] The following reference Figures 1 to 14 This invention describes a product analysis method, a product analysis apparatus, an electronic device, and a readable storage medium according to some embodiments of the present invention.

[0077] Example 1:

[0078] like Figure 1 As shown, according to an embodiment of the present invention, a product analysis method is proposed, the method comprising:

[0079] Step 102: Obtain the consultation record information of the target user within a preset time period;

[0080] Step 104: Based on the consultation record information, identify at least one second product related to the first product;

[0081] Step 106: Based on the consultation record information and N preset consultation information, determine the first consultation probability of the first product and the second consultation probability of the second product;

[0082] Step 108: Based on the first consultation probability and the second consultation probability, determine the difference information and homogeneity information of the first product and the second product from N preset consultation information.

[0083] In this embodiment, for a specific product (i.e., the first product), the target users and times that initiate inquiries about it on the e-commerce platform are obtained. Based on the time of the inquiries, the inquiry record information of the target users within a preset time period (e.g., 24 hours before and after the inquiry time) is obtained. In the inquiry record information, at least one potential competitor (i.e., the second product) that is a product of the same category as the specific product is identified.

[0084] Furthermore, based on N consultation intentions (i.e., preset consultation information), statistical analysis is performed on the consultation record information to obtain the consultation intentions of target users regarding specific products. Using these consultation intentions, the degree of preference of target users for consultation intentions regarding specific products (i.e., the first consultation probability) is analyzed. Additionally, the degree of preference of target users for consultation intentions regarding potential competitors is analyzed using these consultation intentions.

[0085] Furthermore, the analysis examines the target users' preferences for specific products and their preferences for potential competitors. From N inquiry intentions, the analysis identifies the unique features of the product that cannot be replaced by potential competitors (i.e., differentiating information) and the similarities that the product can be replaced by potential competitors (i.e., homogeneous information) for the target users.

[0086] This application statistically analyzes consultation session data of the same user regarding the same type of product within the same time period to obtain the user's consultation intent for specific products and potential competitors. Using this consultation intent, it determines the user's preference for specific products and potential competitors, further analyzes the user's emotional trajectory towards these products, calculates the competitive relationship between products, and identifies the differences and similarities between specific products and potential competitors, thus achieving the discovery of competitive relationships between products. Compared to existing technologies that rely on acquiring large amounts of data from different users' access and search reviews to identify competitors, this application constructs a product comparison combination consistent with the same user's decision-making behavior, achieving consistency in competitor comparison and effectively improving the accuracy of comparing specific products with potential competitors.

[0087] It should be noted that the N preset consultation messages are consultation intentions for the product summarized in advance based on the model. They are used to determine in advance the user's intention to make a certain response to the product, representing a preparatory state before taking action in the consumer behavior process. Among them, consultation intentions can be pre-sales questions, such as product introductions and preferential policies, or post-sales questions, such as logistics tracking and how to return or exchange goods.

[0088] Example 2:

[0089] like Figure 2 As shown, according to an embodiment of the present invention, a product analysis method is proposed, the method comprising:

[0090] Step 202: Obtain the consultation record information of the target user within a preset time period;

[0091] Step 204: Obtain the category information of the first product;

[0092] Step 206: Based on the category information, identify at least one second product related to the first product in the consultation record information;

[0093] Step 208: Based on the consultation record information and N preset consultation information, determine the first consultation probability of the first product and the second consultation probability of the second product;

[0094] Step 210: Based on the first consultation probability and the second consultation probability, determine the difference information and homogeneity information of the first product and the second product from N preset consultation information.

[0095] In this embodiment, based on the category information of a specific product, the product combination within the same product category that is closest to the category information of the specific product is obtained. Then, at least one product from the product combination appearing in the consultation record information is selected as a potential competitor. By statistically analyzing the consultation record information of the same user regarding the same product category within a preset time period, potential competitors related to a specific product are identified, improving the accuracy of potential competitor selection, reducing the impact of irrelevant products on the analysis of a specific product, and thus improving the accuracy of product analysis.

[0096] Specifically, product category information includes brand, category, price, weight, color, size, material, place of origin, packaging, and logistics, etc., which are not limited in this application.

[0097] In a specific embodiment, if the featured product is the Huawei P40, then based on the most detailed category information in the category information such as the brand, category, price, weight, and size of the Huawei P40, it is determined that the related products of the Huawei P40 include the iPhone 11, Xiaomi 10, and OPPO Reno3 Pro. Consultation sessions in which the target user inquires about the Huawei P40 within a preset time period are obtained. Based on these consultation sessions, statistical analysis is performed on the target user's consultation dialogues about related products, and it is determined that the related product inquired by the target user within the preset time period is the iPhone 11. Therefore, the iPhone 11 is a potential competitor of the specific product Huawei P40.

[0098] Example 3:

[0099] like Figure 3 As shown, according to an embodiment of the present invention, a product analysis method is proposed, the method comprising:

[0100] Step 302: Obtain the consultation record information of the target user within a preset time period;

[0101] Step 304: Based on the consultation record information, identify at least one second product related to the first product;

[0102] Step 306: Obtain the first consultation record information related to the first product from the consultation record information;

[0103] Step 308: Determine the first frequency of occurrence of each preset consultation message from the first consultation record information;

[0104] Step 310: Calculate the first consultation probability based on the first frequency and the first formula;

[0105] Step 312: Determine the second consultation probability of the second product based on the consultation record information and N preset consultation information;

[0106] Step 314: Based on the first consultation probability and the second consultation probability, determine the difference information and homogeneity information of the first product and the second product from N preset consultation information.

[0107] In this embodiment, the consultation record information is statistically analyzed to obtain the consultation records of the target user for the specific product (i.e., the first consultation record information). Using the consultation record information, all consultation intentions of the target user for the specific product are determined, the frequency of each consultation intention is calculated, and the sum of the frequencies of all consultation intentions is calculated.

[0108] Furthermore, the probability of each consultation intent is calculated using the formula, which represents the target user's preference level for each consultation intent related to a specific product. By calculating the target user's preference level for a specific product, the user's attention to characteristic products is determined, enabling analysis of the target user's emotional trajectory towards specific products and driving product optimization and upgrades.

[0109] Specifically, the first formula is:

[0110]

[0111] Wherein, e refers to the first product, and I refers to the first product. e For all preset consultation information appearing in the first consultation record information, the above I ek For the kth preset consultation information appearing in the first consultation record information, the above A ek Let F(I) be the probability of the k-th preset consultation information appearing in the first consultation record information. ek ) represents the frequency of the k-th preset consultation message. The sum of the frequencies of all preset consultation information, k∈{1,2,3,…,N}.

[0112] In a specific embodiment, if the effective consultation intentions summarized in advance based on the model are five types: product image quality, product sound effects, preferential policies, logistics inquiry, and how to return or exchange goods, and through statistical analysis of consultation sessions for a specific product, it is determined that the consultation intentions appearing in the consultation sessions for that specific product include three types: product image quality, preferential policies, and how to return or exchange goods. Among them, the frequency of product image quality appears twice, the frequency of preferential policies appears three times, and the frequency of how to return or exchange goods appears once. It is determined that the sum of the frequencies of all corresponding consultation intentions for that specific product is 6 times, the preference degree of product image quality is 2 / 6, the preference degree of product sound effects is 0, the preference degree of preferential policies is 3 / 6, the preference degree of logistics inquiry is 0, and the preference degree of how to return or exchange goods is 1 / 6.

[0113] Example 4:

[0114] like Figure 4 As shown, according to an embodiment of the present invention, a product analysis method is proposed, the method comprising:

[0115] Step 402: Obtain the consultation record information of the target user within a preset time period;

[0116] Step 404: Based on the consultation record information, identify at least one second product related to the first product;

[0117] Step 406: Determine the first consultation probability of the first product based on the consultation record information and N preset consultation information.

[0118] Step 408: Obtain the second consultation record information related to the second product from the consultation record information;

[0119] Step 410: Determine the second frequency of occurrence of each preset consultation message from the second consultation record information;

[0120] Step 412: Calculate the second consultation probability based on the second frequency and the second formula;

[0121] Step 414: Based on the first consultation probability and the second consultation probability, determine the difference information and homogeneity information of the first product and the second product from N preset consultation information.

[0122] In this embodiment, the consultation record information is statistically analyzed to obtain the consultation records of the target user regarding potential competitors (i.e., the second consultation record information). Using the consultation records, all consultation intentions of the target user regarding potential competitors are determined, the frequency of each consultation intention is calculated, and the sum of the frequencies of all consultation intentions is calculated.

[0123] Furthermore, the probability of each consultation intent occurring is calculated using the formula, which represents the target user's preference for consultation intent related to potential competitors. By calculating the target user's preference for consultation intent related to potential competitors, the level of attention the target user pays to potential competitors is determined, thereby enabling the analysis of the target user's emotional trajectory towards potential competitors.

[0124] Specifically, the second formula is:

[0125]

[0126] Wherein, j is the second product, and i is the third product. j For all preset consultation information appearing in the second consultation record information, the above i jk For the kth preset consultation information appearing in the second consultation record information, the above a jk Let F(i) be the probability of the k-th preset consultation information appearing in the second consultation record information. jk ) represents the frequency of the k-th preset consultation message. The sum of the frequencies of all preset consultation information, k∈{1,2,3,…,N}.

[0127] In a specific embodiment, if the effective consultation intentions summarized in advance based on the model are five types: product image quality, product sound effects, preferential policies, logistics tracking, and return / exchange methods, and through statistical analysis of consultation records of potential competitors, it is determined that the consultation intentions appearing are product image quality and preferential policies. Among them, the frequency of product image quality appears once, and the frequency of preferential policies appears four times. The sum of the frequencies of all preset consultation information is determined to be five times. The preference level of product image quality is 1 / 5, the preference level of product sound effects is 0, the preference level of preferential policies is 4 / 5, the preference level of logistics tracking is 0, and the preference level of return / exchange is 0.

[0128] Example 5:

[0129] like Figure 5 As shown, according to an embodiment of the present invention, a product analysis method is proposed, the method comprising:

[0130] Step 502: Obtain the consultation record information of the target user within a preset time period;

[0131] Step 504: Based on the consultation record information, identify at least one second product related to the first product;

[0132] Step 506: Based on the consultation record information and N preset consultation information, determine the first consultation probability of the first product and the second consultation probability of the second product;

[0133] Step 508: Determine the difference between the first product and the second product based on the first consultation probability and the second consultation probability;

[0134] Step 510: Based on the difference value, the first consultation probability, and the second consultation probability, determine the difference information and homogeneity information of the first product and the second product from N preset consultation information.

[0135] In this embodiment, the difference between the specific product and the potential competitor is determined based on the target user's preference for consulting about a specific product and their preference for consulting about potential competitors. Using these difference values, the target user's preference for the specific product, and their preference for the potential competitor, the analysis identifies the points where the specific product cannot be replaced by the potential competitor, as well as the points where the specific product can be replaced by the potential competitor. Compared to existing methods that superficially analyze competitors based on the behavioral chains of all users, this application uses the mindset and motivation of the same user to conduct a more precise analysis of competitors, achieving consistency in competitor comparison and improving the accuracy of obtaining differences and similarities between products.

[0136] Example 6:

[0137] like Figure 6 As shown, according to an embodiment of the present invention, a product analysis method is proposed, the method comprising:

[0138] Step 602: Obtain the consultation record information of the target user within a preset time period;

[0139] Step 604: Based on the consultation record information, identify at least one second product related to the first product;

[0140] Step 606: Based on the consultation record information and N preset consultation information, determine the first consultation probability of the first product and the second consultation probability of the second product.

[0141] Step 608: Determine if the probability of the second consultation is equal to 0. If yes, proceed to step 610; otherwise, proceed to step 612.

[0142] Step 610, the difference value is 0;

[0143] Step 612: Calculate the difference value according to the third formula;

[0144] Step 614: Based on the difference value, the first consultation probability, and the second consultation probability, determine the difference information and homogeneity information of the first product and the second product from N preset consultation information.

[0145] In this embodiment, the consultation records corresponding to potential competitors are statistically analyzed. If the frequency of a certain consultation intention is 0, it means that the target user's preference for the consultation intention of the potential competitor is 0, and the difference between the specific product and the potential competitor for the consultation intention is 0.

[0146] Furthermore, if the frequency of a particular consultation intent is not zero, meaning the target user's preference for that consultation intent regarding a potential competitor is not zero, the ratio of the target user's preference for that consultation intent regarding a specific product to their preference for that consultation intent regarding a potential competitor is calculated. This ratio represents the difference between the specific product and the potential competitor, thus determining the difference values ​​for all consultation intents. By calculating the difference values ​​for the same consultation intent across products, the accuracy of the comparative analysis between the specific product and potential competitors is effectively improved.

[0147] Specifically, the third formula is:

[0148]

[0149] Among them, the above L ejk The difference between the first product and the second product, A above. ek As the probability of the first consultation, the above a jk This represents the probability of the second consultation.

[0150] Example 7:

[0151] like Figure 7 As shown, according to an embodiment of the present invention, a product analysis method is proposed, the method comprising:

[0152] Step 702: Obtain the consultation record information of the target user within a preset time period;

[0153] Step 704: Based on the consultation record information, identify at least one second product related to the first product;

[0154] Step 706: Based on the consultation record information and N preset consultation information, determine the first consultation probability of the first product and the second consultation probability of the second product;

[0155] Step 708: Determine the difference between the first product and the second product based on the first consultation probability and the second consultation probability;

[0156] Step 710: From N preset consultation information, determine the first consultation information that meets the first preset condition;

[0157] Step 712: Use the first consultation information as the difference information between the first product and the second product.

[0158] In this embodiment, the difference between any consultation intention and the first threshold is calculated. If the absolute value of the difference is less than the difference threshold (i.e., the second threshold), the frequency of the consultation intention in the consultation records of the characteristic product is greater than the frequency threshold (i.e., the third threshold), and the frequency of the consultation intention in the consultation records of the potential product is greater than the frequency threshold, it indicates that for the target user, the consultation intention is a difference point that the specific product cannot be replaced by the potential product.

[0159] The first preset condition is that the absolute value of the difference between the difference value and the first threshold is greater than the second threshold, the first consultation probability is greater than the third threshold, and the second consultation probability is greater than the third threshold.

[0160] It should be noted that the first threshold is 1. The absolute value of the difference value minus 1 is used as the judgment value for difference points and homogeneity points. The further the judgment value corresponding to any consultation intention is from 0, the greater the difference in the user's preference for that consultation intention regarding the featured product and the potential product. When the judgment value is greater than the difference threshold, and the frequency of the consultation intention for both the featured product and the potential competitor is greater than the frequency threshold, the consultation intention is determined to be a difference point between the specific product and the potential competitor. By limiting the size of the difference threshold and the homogeneity threshold, the consistency of the judgment logic is ensured, effectively improving the accuracy of product analysis.

[0161] Furthermore, the values ​​of the second and third thresholds are set according to the actual situation, such as the number of homogeneous and differential information required, and are not limited in this application.

[0162] Example 8:

[0163] like Figure 8 As shown, according to an embodiment of the present invention, a product analysis method is proposed, the method comprising:

[0164] Step 802: Obtain the consultation record information of the target user within a preset time period;

[0165] Step 804: Based on the consultation record information, identify at least one second product related to the first product;

[0166] Step 806: Based on the consultation record information and N preset consultation information, determine the first consultation probability of the first product and the second consultation probability of the second product;

[0167] Step 808: Determine the difference between the first product and the second product based on the first consultation probability and the second consultation probability;

[0168] Step 810: From N preset consultation information, determine the second consultation information that meets the second preset condition;

[0169] Step 812: Treat the second consultation information as homogeneous information of the first and second products.

[0170] In this embodiment, the difference between any consultation intent and a first threshold is calculated. If the absolute value of the difference is greater than a homogeneity threshold (i.e., a fourth threshold), the frequency of the consultation intent appearing in the consultation records of the featured product is greater than a frequency threshold, and the frequency of its appearance in the consultation records of the potential product is also greater than a frequency threshold. This indicates that, for the target user, the consultation intent represents a homogeneous point where the specific product can be replaced by the potential product. This application determines the differences and homogeneities between products based on the same comparison standard, achieving consistency in competitor comparison and improving the accuracy of selecting differences and homogeneities.

[0171] The second preset condition is that the absolute value of the difference between the difference value and the first threshold is less than the fourth threshold, the first consultation probability is greater than the third threshold, and the second consultation probability is greater than the third threshold; wherein, 0 < fourth threshold < second threshold < 1.

[0172] It should be noted that the first threshold is 1. The absolute value of the difference value minus 1 is used as the judgment value for difference points and homogeneity points. The closer the judgment value corresponding to any consultation intention is to 0, the more similar the user's preference for that consultation intention is for the specific product and potential competitors. When the judgment value is less than the homogeneity threshold, and the frequency of the consultation intention is greater than the frequency threshold for both the specific product and potential competitors, the consultation intention corresponding to that judgment value is determined to be a homogeneity point between the specific product and potential competitors.

[0173] Among them, 0 < homogeneity threshold < difference threshold < 1. By limiting the difference threshold and homogeneity threshold, the consistency of the judgment logic is ensured, and the accuracy of product analysis is improved.

[0174] Furthermore, the values ​​of the third and fourth thresholds are set according to the actual situation, such as the number of homogeneous and differential information required, and are not limited in this application.

[0175] Example 9:

[0176] like Figure 9 As shown, according to an embodiment of the present invention, a product analysis method is proposed, the method comprising:

[0177] Step 902: Obtain the consultation record information of the target user within a preset time period;

[0178] Step 904: Based on the consultation record information, identify at least one second product related to the first product;

[0179] Step 906: Based on the consultation record information and N preset consultation information, determine the first consultation probability of the first product and the second consultation probability of the second product;

[0180] Step 908: Based on the first consultation probability and the second consultation probability, determine the difference information and homogeneity information of the first product and the second product from N preset consultation information;

[0181] Step 910: When there are multiple second products, determine the similarity between the first product and each second product based on the first consultation probability and the second consultation probability.

[0182] Step 912: Identify a second product from multiple second products whose similarity is greater than a similarity threshold.

[0183] In this embodiment, when there are multiple potential competitors, the similarity between the specific product and each potential competitor's consultation intent is determined based on the target user's preference for the specific product and the potential competitor's consultation intent. Multiple similarity values ​​are then ranked; a higher similarity indicates a smaller difference in the target user's consultation intent between the two products, meaning the potential competitor and the specific product are more similar based on the target user's preferences. Conversely, a lower similarity indicates a greater difference in the target user's consultation intent between the two products, meaning the potential competitor and the specific product are less similar based on the target user's preferences. By comparing the similarity of the target user's consultation intent between the specific product and the potential competitor, the difference in the target user's attention to the two products is determined, improving the accuracy of the analysis of the degree of competitiveness between products.

[0184] In a specific embodiment, statistical analysis of consultation records for TV A reveals that TVs B and C are its competitors and have similar competitive relationships. However, target users pay attention to A because of its picture quality, and they also pay attention to B because of its picture quality, while C is attracted by its sound effects. As a result, B is more closely competitive with A than C. The similarity between target users and TVs B and C is calculated and analyzed. The higher the similarity, the greater the degree of homogeneous competitiveness between the two products.

[0185] Example 10:

[0186] like Figure 10 As shown, according to a specific embodiment of the present invention, a product analysis method is proposed, the method comprising:

[0187] Step 1002: Obtain consultation record information of the second product combination within a preset time period based on the first product;

[0188] Step 1004: Obtain preset consultation information for the first and second products;

[0189] Step 1006: Preset consultation information statistics and obtain the consultation probability vector of the target user;

[0190] Step 1008: Obtain similarity;

[0191] Step 1010: Obtain difference information and homogeneity information.

[0192] In this embodiment, such as Figure 11 and Figure 12 As shown, based on a specific product, consultation records of potential competitor combinations are obtained within a preset time period. For a specific product, such as product P... e Obtain the target user and time T who initiated the inquiry, and based on time T, find the four-tuple of inquiries from all other products in the same category within the time interval t before and after time T, i.e., the time interval [Tt, T+t], for example, within 24 hours before and after time T. <P e p j S e s j >, where P e For the specific product being consulted, p j For the corresponding product in the consultation, S e For consulting on specific product P e Record information, s j For inquiries about the corresponding product p j Record information.

[0193] Furthermore, based on product portfolio <P e ,p j From all the recorded information pairs, obtain their corresponding preset consultation information, that is, their corresponding intent, and construct the corresponding six-tuple. <P e p j S e s j I e i j >, where I e For consulting on specific product P e Record information S e The corresponding intent, i j For inquiries about the corresponding product p j Record information s j Based on the user's intent to inquire about different products within the same category in a similar timeframe, potential competitors can be identified.

[0194] Furthermore, based on the statistical distribution of consultation intentions for a specific product and potential competitors, a preference vector is constructed, and based on this preference vector, the similarity relationship between the specific product and potential competitors is calculated. Specifically, such as... Figure 11 As shown, for <P e p j All record information e s j All intentions e i​​j > Construct their respective consultation intent preference vectors {A e}={A ek |k=1,2,3,…,n} and {a j}={a jk |k=1, 2, 3,...,n}. Specifically, A ek =F(I ek ) / ΣF(I e ), where A ek For consulting on specific product P e The preference of the kth consultation intention, A e For consulting on specific product P e All consultation intentions preferences, I ek For consulting on specific product P e The kth consultation intent, F(I) e For consulting on specific product P e The kth consultation intent I ek The frequency of occurrence, ΣF(I e For consulting on specific product P e The sum of the frequencies of all consultation intentions, further, a jk =F(i jk ) / ΣF(i j ), where a jk For inquiries about the corresponding product p j The preference of the kth consultation intention, a j For inquiries about the corresponding product p j All consultation intentions preferences, i jk For inquiries about the corresponding product p j The kth consultation intent, F(i) jk (For consulting on the corresponding product p) j The kth consultation intent i jk The frequency of occurrence, ΣF(i) j ) represents the corresponding product p j The sum of the frequencies of all consultation intentions.

[0195] Furthermore, such as Figure 11 As shown, based on {A e} and {a j} Calculate the similarity cosine coefficients, specifically, Among them, R ejk Let P be the similarity cosine coefficient between the k-th consultation intent for a specific product and its corresponding product. e Obtain its corresponding product p j The cosine coefficient vector {R ej}={R ejk |k=1,2,3,…,n}, where Rej For P e and p j , the cosine similarity coefficient of all consultation intentions. Sort R ejk , the higher R ejk , the more similar the two products are based on user preferences, and vice versa.

[0196] Furthermore, based on the preference vectors of specific products and corresponding products, construct a difference value, and based on the difference value and the threshold of frequency, obtain difference points and homogeneous points. Specifically, as Figure 12 shown, for P e and p j , based on their corresponding preference vectors {A e} and {a j}, construct the corresponding difference value vector {L ej} = {L ejk |k = 1, 2, 3,..., n}, where L ej is the difference value of all consultation intentions of P e and p j , and L ejk is the difference value of the k-th consultation intention of P e and p j . For L ejk , when a jk is equal to 0, L ejk is equal to 0; when a jk is greater than 0,

[0197] Furthermore, based on the difference value, construct a similarity judgment vector {U ej} = {U ejk = |L ejk - 1||k = 1, 2, 3,..., n}, where U ej is the similarity judgment value of all consultation intentions of P e and p j , and U ejk is the similarity judgment value of the k-th consultation intention of P e and p j . For each element U ejk in the vector, if U ejk > Td, and F(I ek )> Tf, F(i jk )> Tf, then the consultation intention I ek is a difference point between P e and p j ; if U ejk < Ts, and F(I ek )> Tf,, F(i jk )> Tf, then the consultation intention Iek For P e and p j The homogeneous points are defined by Td, Ts, and Tf, where Td is the difference threshold, Ts is the homogeneity threshold, and Tf is the frequency threshold, and 1>Td>Ts>0.

[0198] Compared to existing technologies that rely on accessing and searching to acquire large amounts of data and then use that data to identify competitors, this approach only superficially identifies competitors based on user behavior chains. It fails to capture users' mindsets and motivations for more accurate competitor identification. Furthermore, while reviews reveal attitudes towards different competitor characteristics and user preferences, these reviews, while based on actual user experience, cannot guarantee that the users are the same across different products, and the comparison criteria may vary. Therefore, consistency in competitor comparison is lacking. This application, however, utilizes online consultation session data from the same user regarding the same product category on e-commerce platforms within the same timeframe. By obtaining the intent behind the consulted product combinations, it constructs product comparison combinations consistent with the user's decision-making behavior. Based on intent statistics, it generates similarity data for different products within the same category in consultations, thereby calculating the competitive relationship between products and identifying points of similarity and difference to achieve competitor discovery.

[0199] Example 11:

[0200] like Figure 13 As shown, according to an embodiment of the second aspect of the present invention, a product analysis device 1300 is provided, comprising: an acquisition module 1302, configured to acquire consultation record information of a target user within a preset time period; a first determination module 1304, configured to determine at least one second product related to a first product based on the consultation record information; a second determination module 1306, configured to determine a first consultation probability of the first product and a second consultation probability of the second product based on the consultation record information and N preset consultation information, wherein N is an integer greater than 0; and a third determination module 1308, configured to determine difference information and homogeneity information of the first product and the second product from the N preset consultation information based on the first consultation probability and the second consultation probability.

[0201] In this embodiment, the product analysis device 1300 includes an acquisition module 1302, a first determination module 1304, a second determination module 1306, and a third determination module 1308.

[0202] Specifically, the acquisition module 1302 is configured to acquire the target user and time of inquiries made to a specific product (i.e., the first product) on the e-commerce platform, and based on the time of the inquiries, acquire the inquiry record information of the target user within a preset time period (e.g., 24 hours before and after the inquiry time).

[0203] Furthermore, the first determining module 1304 is configured to identify at least one potential competitor (i.e., a second product) that is a product of the same category as the specific product in the consultation record information.

[0204] Furthermore, the second determining module 1306 is configured to perform statistical analysis on the consultation record information based on N consultation intentions (i.e., preset consultation information), obtain the consultation intentions of target users for consulting about specific products appearing in the consultation record information, analyze the degree of preference of target users for consulting about specific products (i.e., the first consultation probability) using the consultation intentions, and analyze the degree of preference of target users for consulting about potential competitors appearing in the consultation record information (i.e., the second consultation probability) using the consultation intentions.

[0205] Furthermore, the third determining module 1308 is configured to analyze the target user's preference for a specific product and their preference for potential competitors. From N consultation intentions, it determines the differences (i.e., difference information) that the featured product cannot be replaced by potential competitors for the target user, and the homogeneous points (i.e., homogeneous information) that the featured product can be replaced by potential competitors.

[0206] This application statistically analyzes consultation session data of the same user regarding the same type of product within the same time period to obtain the user's consultation intent for specific products and potential competitors. Using this consultation intent, it determines the user's preference for specific products and potential competitors, further analyzes the user's emotional trajectory towards these products, calculates the competitive relationship between products, and identifies the differences and similarities between specific products and potential competitors, thus achieving the discovery of competitive relationships between products. Compared to existing technologies that rely on acquiring large amounts of data from different users' access and search reviews to identify competitors, this application constructs a product comparison combination consistent with the same user's decision-making behavior, achieving consistency in competitor comparison and effectively improving the accuracy of comparing specific products with potential competitors.

[0207] Example 12:

[0208] According to an embodiment of a third aspect of the present invention, an electronic device is provided, comprising an analysis apparatus for the product proposed in the second aspect. Therefore, this electronic device possesses all the beneficial effects of the analysis apparatus for the product proposed in the second aspect, and to avoid repetition, further details will not be provided.

[0209] Example 13:

[0210] like Figure 14As shown, according to an embodiment of the fourth aspect of the present invention, an electronic device 1400 is provided, comprising: a memory 1402 storing a program or instructions; and a processor 1404 connected to the memory 1402, configured to implement the analysis method of the product proposed in the first aspect when executing the program or instructions. Therefore, this electronic device possesses all the beneficial effects of the analysis method of the product proposed in the first aspect, which will not be elaborated further here.

[0211] Example 14:

[0212] According to a fifth aspect of the present invention, a readable storage medium is provided on which a program or instructions are stored, which, when executed by a processor, perform the analysis method of the product proposed in the first aspect. Therefore, this readable storage medium possesses all the beneficial effects of the analysis method of the product proposed in the first aspect, and to avoid repetition, further details are omitted.

[0213] In this invention, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0214] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods of the various embodiments of this application.

[0215] In the description of this specification, the terms "one embodiment," "some embodiments," "specific embodiment," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0216] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method of analyzing a product, characterized by, The method comprises the following steps: obtaining a target user and a time of initiating consultation on a first product on an e-commerce platform, and obtaining consultation record information of the target user within a preset time period based on the time of initiating consultation; determining at least one second product related to the first product according to the consultation record information and category information of the first product; determining a first consultation probability of the first product and a second consultation probability of the second product according to the consultation record information and N preset consultation information, wherein N is an integer greater than 0; determining difference information and homogeneous information of the first product and the second product from N preset consultation information according to the first consultation probability and the second consultation probability; The step of determining the difference information and the homogeneous information of the first product and the second product from N preset consultation information according to the first consultation probability and the second consultation probability specifically comprises: determining a difference value of the first product and the second product according to the first consultation probability and the second consultation probability; determining the difference information and the homogeneous information of the first product and the second product from N preset consultation information according to the difference value, the first consultation probability and the second consultation probability; The step of determining the difference value of the first product and the second product according to the first consultation probability and the second consultation probability specifically comprises: based on the second consultation probability being equal to 0, the difference value is 0; based on the second consultation probability not being equal to 0, calculating the difference value according to a third formula; The third formula is: wherein the above L ejk is the difference value of the first product and the second product, the above A ek is the first consulting probability, the above a jk is the second consulting probability.

2. The method of analysis of a product according to claim 1, characterized in that, The step of determining at least one second product corresponding to the first product according to the consultation record information specifically comprises: obtaining category information of the first product; determining at least one second product related to the first product in the consultation record information according to the category information.

3. The method of analysis of a product according to claim 1, characterized in that, The step of determining the first consultation probability of the first product according to the consultation record information and N preset consultation information specifically comprises: obtaining first consultation record information related to the first product in the consultation record information; determining a first frequency of occurrence of each preset consultation information from the first consultation record information; calculating the first consultation probability according to the first frequency and a first formula; The first formula is: wherein, the above e is the first product, the above I e is all the preset consultation information appearing in the first consultation record information, the above I ek is the kth preset consultation information appearing in the first consultation record information, the above A ek is the probability of the kth preset consultation information appearing in the first consultation record information, the above F(I ek ) is the frequency of the kth preset consultation information appearing, the above is the sum of the frequencies of all the preset consultation information appearing, k ∈ {1, 2, 3, …, N}.

4. The method of analysis of a product according to claim 1, characterized in that, The step of determining the second consultation probability of the second product according to the consultation record information and N preset consultation information specifically comprises: obtaining second consultation record information related to the second product in the consultation record information; determining a second frequency of occurrence of each preset consultation information from the second consultation record information; calculating the second consultation probability according to the second frequency and a second formula; The second formula is: wherein, the above j is the second product, the above i j is all the preset consultation information appearing in the second consultation record information, the above i jk is the kth preset consultation information appearing in the second consultation record information, the above a jk is the probability of the kth preset consultation information appearing in the second consultation record information, the above F(i jk ) is the frequency of the kth preset consultation information appearing, the above is the sum of the frequencies of all the preset consultation information appearing, k∈{1, 2, 3, …, N}.

5. The method of analysis of a product according to claim 1, characterized in that, The step of determining the difference information and the homogeneous information of the first product and the second product from N preset consultation information according to the difference value, the first consultation probability and the second consultation probability specifically comprises: determining first consultation information meeting a first preset condition from N preset consultation information; determine the second consultation information from the N preset consultation information, the second preset condition being met; determine the second consultation information from the N preset consultation information, the second preset condition being met; determine the second consultation information from the N preset consultation information, the second preset condition being met; wherein 0 < the fourth threshold value < the second threshold value < 1.

6. The method of analysis of a product according to claim 1, characterized in that, when the number of the second products is a plurality, further comprising: determine the similarity between the first product and each of the second products according to the first consultation probability and the second consultation probability; determine the second product from the plurality of second products, the similarity between the first product and the second product being greater than a similarity threshold value.

7. An apparatus for analyzing a product, characterized by comprising: an acquisition module, configured to acquire a target user and a time of initiating a consultation on a first product on an e-commerce platform, and acquire consultation record information of the target user within a preset time period based on the time of initiating the consultation; a first determination module, configured to determine at least one second product related to the first product according to the consultation record information and category information of the first product; a second determination module, configured to determine a first consultation probability of the first product and a second consultation probability of the second product according to the consultation record information and N preset consultation information, N being an integer greater than 0; a third determination module, configured to determine difference information and homogeneity information of the first product and the second product from the N preset consultation information according to the first consultation probability and the second consultation probability; the step of determining the difference information and the homogeneity information of the first product and the second product from the N preset consultation information according to the first consultation probability and the second consultation probability, specifically comprising: determining a difference value of the first product and the second product according to the first consultation probability and the second consultation probability; determining the difference information and the homogeneity information of the first product and the second product from the N preset consultation information according to the difference value, the first consultation probability and the second consultation probability; the step of determining the difference value of the first product and the second product according to the first consultation probability and the second consultation probability, specifically comprising: based on the second consultation probability being equal to 0, the difference value being 0; based on the second consultation probability not being equal to 0, calculating the difference value according to a third formula; the third formula being: wherein the above L ejk is the difference value of the first product and the second product, the above A ek is the first consulting probability, the above a jk is the second consulting probability.

8. An electronic device, comprising: comprising: the product analysis device of claim 7.

9. An electronic device, comprising: comprising: a memory, the memory storing a program or instructions; a processor connected with the memory, the processor executing the product analysis method of any one of claims 1 to 6 when executing the program or instructions.

10. A readable storage medium, on which a program or instructions are stored, characterized in that, The program or the instruction realizes the steps of the analysis method of the product as claimed in any one of claims 1 to 6 when executed by the processor.

Images