Product pain point identification method, device, electronic device and storage medium
By matching and grading the pain points of negative emotional product review texts, identifying core, public, emergency and edge pain points, solving the problem of poor pain point recognition effect in the existing technology, achieving more accurate pain point recognition and decision-making support, and enhancing enterprises market competitiveness.
Patent Information
- Application Number
- CN202410109913.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-26
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-01-26
AI Technical Summary
The existing technology has poor effect in identifying product pain points, and it is impossible to accurately identify the pain points in consumer feedback, making it difficult for enterprises to optimize products and improve competitiveness.
Using a pain point label library method, pain point matching is performed on negative emotions, and grading it through universality and urgency indicators to identify core, public, emergency and edge pain points, and improve the accuracy and readability of pain point recognition. .
It improves the accuracy and decision-making efficiency of product pain points identification, helps enterprises optimize resource allocation, expand new market opportunities, and enhance product competitiveness.
Smart Images

Figure CN117649255B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of artificial intelligence technology, and specifically to a product pain point identification method, device, electronic device and storage medium. Background Art
[0002] Pain points are the difficulties, problems or dissatisfactions that consumers or users encounter when using a product. These pain points may include product quality, price, function, user experience, after-sales service, etc. Understanding consumers' pain points can help companies better meet consumer needs and improve product competitiveness.
[0003] Existing technologies for obtaining consumer feedback are limited to qualitative and quantitative surveys of small samples. With the continuous development of Internet big data technology, consumer feedback can be obtained from various e-commerce platforms more timely than in the past. Consumer feedback corpus has the characteristics of long text, context dependence, and informal wording. At present, the effect of identifying product pain points in consumer feedback corpus is poor, which is not conducive to improving the competitiveness of enterprise products. Summary of the invention
[0004] The present application aims to solve the technical problem that the current product pain point identification effect is poor. To this end, the present application proposes a product pain point identification method, which can improve the recognition accuracy of product pain points.
[0005] This application also proposes a product pain point identification device.
[0006] The present application also proposes an electronic device and a non-transitory computer-readable storage medium.
[0007] A product pain point identification method according to an embodiment of the first aspect of the present application includes:
[0008] Based on the pain point tag library, the target product review text with negative emotions within a preset time period is matched with pain points to obtain the target product pain point matching result; the pain point tag library includes an industry general tag library, an industry characteristic tag library and an inapplicable tag library; the industry general tag library is a tag library determined based on the general pain point words of the industry to which the target product belongs; the industry characteristic tag library is a tag library determined based on the characteristic pain point words of the industry to which the target product belongs; the inapplicable tag library is a tag library determined based on the inapplicable pain point words of the target product in the general pain point words of the industry general tag library;
[0009] Based on the universality index and the urgency index of each pain point, the pain point matching results are graded to obtain the pain point identification results of the target product; the universality index of each pain point is determined based on the ratio of the number of each pain point in the pain point matching results to the total number of pain points; the urgency index of each pain point is determined based on the growth rate of the number of each pain point in the pain point matching results compared to the corresponding number of pain points in the previous preset time period.
[0010] The product pain point identification method provided in the embodiment of the present application matches the target product review text with negative emotions with pain point through the pain point tag library, and then grades the pain points based on the universal index and the urgency index, thereby obtaining the target product pain point identification result, which can improve the accuracy of product pain point identification, improve the readability of pain points, and avoid the high cost of manual interpretation and correction. In addition, by introducing universality and urgency to grade pain points and quantify the value of pain points, it is helpful to discover new groups, new scenarios, etc., expand new product opportunities, improve the efficiency of decision-making on pain point processing, and significantly guide the optimization of enterprise pain point resource allocation and improve the competitiveness of enterprise products.
[0011] According to an embodiment of the present application, the pain point matching results are graded based on the universality index of each pain point and the urgency index of each pain point to obtain the target product pain point identification result, including:
[0012] Based on the universality index of each pain point, determine the average universality index of all pain points in the pain point matching result; and based on the urgency index of each pain point, determine the average urgency index of all pain points in the pain point matching result;
[0013] Determine a secondary pain point classification map based on the average of the universality index of all the pain points and the average of the urgency index of all the pain points;
[0014] Based on the positions of the prevalence index of each pain point and the urgency index of each pain point in the secondary pain point classification map, the pain point matching results are graded to obtain the target product pain point identification results; the pain point identification results include core pain points, popular pain points, urgent pain points and marginal pain points; the core pain points represent pain points whose prevalence index is greater than or equal to the average of the prevalence indexes of all pain points, and whose urgency index is greater than the average of the urgency indexes of all pain points; the popular pain points represent pain points whose prevalence index is greater than or equal to the average of the prevalence indexes of all pain points, and whose urgency index is less than or equal to the average of the urgency indexes of all pain points; the urgent pain points represent pain points whose prevalence index is less than the average of the prevalence indexes of all pain points, and whose urgency index is greater than the average of the urgency indexes of all pain points; the marginal pain points represent pain points whose prevalence index is less than the average of the prevalence indexes of all pain points, and whose urgency index is less than or equal to the average of the urgency indexes of all pain points.
[0015] The product pain point identification method provided in the embodiment of the present application determines the secondary pain point classification map through the mean of the universal index and the mean of the urgency index, thereby classifying the pain points and obtaining core pain points, popular pain points, emergency pain points and marginal pain points. By quantifying the value of pain points, it helps to discover new groups of people, new scenarios, etc., expand new product opportunities, further improve the efficiency of pain point processing decision-making, significantly guide the optimization of resource allocation for corporate pain points, and further improve the competitiveness of corporate products.
[0016] According to an embodiment of the present application, the pain point matching of the target product review text with negative emotions within a preset time period is performed based on the pain point tag library to obtain the target product pain point matching result, including:
[0017] Based on the industry general tag library and industry characteristic tag library in the pain point tag library, pain point matching is performed on the target product review text with negative emotions within a preset time period to obtain the initial target product pain point matching result;
[0018] Based on the inapplicable tag library in the pain point tag library, the inapplicable pain point words in the initial target product pain point matching result are filtered to obtain the target product pain point matching result.
[0019] According to an embodiment of the present application, the pain point matching of target product review texts with negative emotions within a preset time period based on the industry general tag library and the industry characteristic tag library in the pain point tag library includes:
[0020] Generate a first vector based on the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library; and generate a second vector based on the target keywords of the target product review text with negative emotions within a preset time period; the target keywords represent feedback on the target product;
[0021] Perform canonical matching on the first vector and the second vector.
[0022] According to one embodiment of the present application, the method further includes:
[0023] It is determined that the similarity between the target keyword and any product pain point word is greater than a similarity threshold, and the target keyword is mapped to the corresponding product pain point word.
[0024] According to one embodiment of the present application, the method further includes:
[0025] Determine that the ratio of the total number of pain points in the target product pain point matching results to the number of target product review texts with negative emotions within a preset time period is less than a preset ratio, and based on the target keywords in the target product review texts with negative emotions within the preset time period, update the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library.
[0026] According to one embodiment of the present application, the method further includes:
[0027] Based on the negative sentiment words of the industry to which the target product belongs, negative sentiment words are matched on the target product review texts within a preset time period to determine the target product review texts with negative sentiment.
[0028] A product pain point identification device according to the second aspect of the present application includes:
[0029] A matching module is used to perform pain point matching on target product review texts with negative emotions within a preset time period based on a pain point tag library to obtain target product pain point matching results; the pain point tag library includes an industry general tag library, an industry characteristic tag library and an inapplicable tag library; the industry general tag library is a tag library determined based on general pain point words of the industry to which the target product belongs; the industry characteristic tag library is a tag library determined based on characteristic pain point words of the industry to which the target product belongs; the inapplicable tag library is a tag library determined based on inapplicable pain point words of the target product in the general pain point words of the industry general tag library;
[0030] A grading module is used to grade the pain points of the pain point matching results based on the prevalence index and the urgency index of each pain point to obtain the pain point identification results of the target product; the prevalence index of each pain point is determined based on the ratio of the number of each pain point in the pain point matching results to the total number of pain points; the urgency index of each pain point is determined based on the growth rate of the number of each pain point in the pain point matching results compared to the corresponding number of pain points in a previous preset time period.
[0031] The product pain point identification device provided in the embodiment of the present application matches the target product review text with negative emotions with pain point through the pain point tag library, and then grades the pain points based on the universal index and the urgency index, thereby obtaining the target product pain point identification result, which can improve the accuracy of product pain point identification, improve the readability of pain points, and avoid the high cost of manual interpretation and correction. In addition, by introducing universality and urgency to grade pain points and quantify the value of pain points, it is helpful to discover new groups, new scenarios, etc., expand new product opportunities, improve the efficiency of pain point processing decision-making, and significantly guide enterprises to optimize resource allocation based on pain points and improve the competitiveness of enterprise products.
[0032] According to an electronic device according to an embodiment of the third aspect of the present application, the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the product pain point identification method as in the embodiment of the first aspect is implemented.
[0033] According to the non-transitory computer-readable storage medium of the fourth aspect embodiment of the present application, a computer program is stored thereon, and when the computer program is executed by a processor, the product pain point identification method as in the first aspect embodiment is implemented.
[0034] The above one or more technical solutions in the embodiments of the present application have at least one of the following technical effects:
[0035] The product pain point identification method provided in the embodiment of the present application matches the target product review text with negative emotions with pain point through the pain point tag library, and then grades the pain points based on the universal index and the urgency index, thereby obtaining the target product pain point identification result, which can improve the accuracy of product pain point identification, improve the readability of pain points, and avoid the high cost of manual interpretation and correction. In addition, by introducing universality and urgency to grade pain points and quantify the value of pain points, it is helpful to discover new groups, new scenarios, etc., expand new product opportunities, improve the efficiency of decision-making on pain point processing, and significantly guide the optimization of enterprise pain point resource allocation and improve the competitiveness of enterprise products.
[0036] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0038] Figure 1 It is a flowchart of a product pain point identification method provided in an embodiment of the present application;
[0039] Figure 2 This is a schematic diagram of different product requirements for different groups of people in different scenarios provided by the embodiments of the present application;
[0040] Figure 3 It is a schematic diagram of a secondary pain point classification map provided in an embodiment of the present application;
[0041] Figure 4 It is a schematic diagram of a pain point identification and value quantification model provided in an embodiment of the present application;
[0042] Figure 5 It is a flowchart of the product pain point identification method provided in the embodiment of the present application in practical application;
[0043] Figure 6 It is a structural schematic diagram of a product pain point identification device provided in an embodiment of the present application;
[0044] Figure 7 It is a schematic diagram of the physical structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0045] Consumers can get feedback on products more timely from various e-commerce platforms than in the past. For example, on a certain e-commerce platform, users gave the following feedback on the toothpaste they used: "The toothpaste is very good. After brushing, the hair on the teeth is very clean, and there is no stinging feeling. The toothpaste is very easy to use. I bought it for my mother. She likes it very much. It is also very easy to use. After using it once, I feel that my teeth have become whiter and my teeth are very clean." The user's feedback corpus on the toothpaste has contextual dependence and is vague. On another e-commerce platform, users gave the following feedback on several toothpastes they used: "Toothpaste 1: It can be used by people with sensitive mouths. It can freshen breath and protect gums while soothing sensitivity; Toothpaste 2: Baking soda toothpaste, mainly removes yellow stains and deep cleans tartar; Toothpaste 3: Jasmine scented, fresh and fragrant, improves tooth stains and tartar; Toothpaste 4: This is the one I bought this year. It is suitable for people who often drink tea and coffee to use this to improve tooth stains." The user's feedback corpus on toothpaste is a long text and has contextual dependence.
[0046] At present, natural language processing technology is mainly used to extract pain points from product feedback corpus, with tag keyword matching as the main method and sentiment tendency as the auxiliary method to judge and select pain points. However, the overall recognition effect is poor, and there are the following shortcomings:
[0047] (1) The pain points extracted are too general or even wrong, and still require a lot of manpower to interpret and correct;
[0048] (2) There are many pain points, but their value cannot be quantified, and there is no direct guidance for pain point optimization decisions;
[0049] (3) Pain point optimization decisions are highly dependent on human judgment, and experience cannot be converted into data accumulation.
[0050] For example: On a certain e-commerce platform, the user's feedback on a power bank is: "The thing feels okay at the moment, but the charging speed is a bit slow, and the logistics is very slow. It took more than a week to receive the goods. It is useless to compensate for late arrival. I don't understand the meaning of its existence." In the feedback, the user is not satisfied with the charging speed, but the description is general; the description of the pain point in logistics is clear, and the pain point is specifically slow logistics, which is reflected in the fact that it took more than a week to receive the goods. Nowadays, technology can generally only analyze the two pain points of slow charging and slow logistics, and the detailed description of the pain points and the reasons for the pain require manual tracing and interpretation. Based on this, the present application proposes a method that can more accurately and specifically identify the pain points in real consumer feedback on social and e-commerce platforms, and quantify and measure the value of the pain points. This method can ensure that the identified pain points are clearer, thereby reducing the cost of manual interpretation and correction; it can also assist in guiding the optimization of pain point classification, thereby improving decision-making efficiency.
[0051] The following is a further detailed description of the implementation of the present application in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but cannot be used to limit the scope of the present application.
[0052] In the description of the embodiments of the present application, it should be noted that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, which are only for the convenience of describing the embodiments of the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the embodiments of the present application. In addition, the terms "first", "second", and "third" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.
[0053] In the description of the embodiments of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the embodiments of the present application can be understood according to specific circumstances.
[0054] In the embodiments of the present application, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.
[0055] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiments of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0056] Figure 1 : is a flow chart of the product pain point identification method provided in the embodiment of the present application. Figure 1 The product pain point identification method provided in the embodiment of the present application may include:
[0057] Step 110: Based on the pain point tag library, pain point matching is performed on the target product review text with negative emotions within a preset time period to obtain the target product pain point matching result; the pain point tag library includes an industry general tag library, an industry characteristic tag library and an inapplicable tag library; the industry general tag library is a tag library determined based on the general pain point words of the industry to which the target product belongs; the industry characteristic tag library is a tag library determined based on the characteristic pain point words of the industry to which the target product belongs; the inapplicable tag library is a tag library determined based on the inapplicable pain point words of the target product in the general pain point words of the industry general tag library;
[0058] Step 120: Based on the universal index and the urgency index of each pain point, the pain point matching results are graded to obtain the target product pain point identification results; the universal index of each pain point is determined based on the ratio of the number of each pain point in the pain point matching results to the total number of pain points; the urgency index of each pain point is determined based on the growth rate of the number of each pain point in the pain point matching results compared to the corresponding number of pain points in the previous preset time period.
[0059] It should be noted that the execution subject of the product pain point identification method provided in the embodiment of the present application may be a computer device, such as a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook or a personal digital assistant (PDA), etc. The following takes the pain point identification system that executes the product pain point identification method as an example of the execution subject to illustrate the product pain point identification method provided in the embodiment of the present application.
[0060] In step 110, the pain point identification system can obtain user feedback corpus on the target product within a preset time period as the target product review text. The target product review text may include positive emotional review text and negative emotional review text, etc., and the review text with positive emotion usually does not include product pain points. The pain point identification system can filter the target product review text with negative emotion in the target product review text. For example, the screening of negative emotional review text can be achieved by building an emotional dictionary or based on machine learning methods such as naive Bayes, maximum entropy, etc. The negative emotional review text may contain target keywords that represent feedback on the target product, such as "very few", "few gifts", "side effects", etc.
[0061] The pain point identification system can match the target product review text with negative emotions within a preset time period based on the pain point tag library, thereby obtaining the target product pain point matching results. In addition to the product pain points, the target product matching results can also include the number of pain points of each product. For example, the pain point identification results of a certain cosmetics can include not only product pain points such as "will not repurchase", "few gifts", "allergic", etc., but also the statistical results of the number of comments on each product pain point.
[0062] This application abandons the current method of extracting negative keywords from consumer feedback and negative reviews, and instead uses trend lifestyle as a guide, splits the large scene according to the level of human needs, and further splits it into sub-scenes according to differences in consumer concepts and product needs, and then determines the pain point tag library. Figure 2As shown in the figure, taking charging demand as an example, the application is specifically constructed based on the crowd-scenario-pain point. It can be seen that different groups have different priorities for the characteristics of charging products. Compared with the existing defects that can only identify general pain points, this application identifies pain points more clearly and specifically, and can clarify the relationship between people / products / time / events / space, etc. related to pain points.
[0063] This application can monitor the industry hot spots, refine the results of continuous tracking of medium- and long-term trends, analyze qualitative trend signals according to external-internal factors, PEST influences, context evolution, etc., and evaluate the trend potential, and then build a pain point label library. Specifically, for the tracked trends, further hierarchical decomposition and segmentation of scenarios and labels can be performed according to Maslow's needs model, domestic and foreign consumption concepts, lifestyles, etc., and classified according to industry characteristics such as crowds, products, venues, time, etc., to build an industry general label library and an industry characteristic label library. Among them, the industry general label library can be determined based on the general pain point words of the industry to which the target product belongs. Labels such as crowds can be used across industries and can be used as general pain point words for the industry general label library. The industry characteristic label library can be determined based on the characteristic pain point words of the industry to which the target product belongs. Labels such as materials have strong industry characteristics and can be used as characteristic pain point words for the industry characteristic label library.
[0064] In addition to the industry-wide tag library and the industry-specific tag library, the pain point tag library can also include an inapplicable tag library. The inapplicable tag library can be determined based on the inapplicable pain point words for the target product in the general pain point words of the industry-wide tag library. For example: For the washing and care industry, dryness can be used as a general pain point word in the industry-wide tag library of the product, and dryness can be used to describe the user's feedback on the use of washing and care products (such as facial cleansers). For people with dry skin, dryness is a product pain point; while for people with oily skin, dryness is not a product pain point. In this case, the inapplicable tag library can be further determined based on the inapplicable pain point words for the target product in the general pain point words of the industry-wide tag library (the dryness of facial cleansers is not applicable to describe the pain points of people with oily skin).
[0065] In step 120, the pain point identification system can determine the prevalence index of each pain point based on the ratio of the number of each pain point in the pain point matching result to the total number of pain points; and determine the urgency index of each pain point based on the growth rate of the number of each pain point in the pain point matching result compared with the corresponding number of pain points in the previous preset time period. For example: the total number of pain points in the pain point matching result in the preset time period is A, among which the number of a certain pain point is a, and the number of the pain point in the previous preset time period is b. Then the prevalence index of the pain point is equal to the number of the pain point in the preset time period / the total number of pain points in the preset time period × 100%, that is, a / A×100%; the urgency index of the pain point is equal to (the number of the pain point in the preset time period – the number of the pain point in the previous preset time period) / the number of the pain point in the previous preset time period × 100%, that is, (ab) / b×100%.
[0066] The pain point identification system can classify each pain point in the pain point matching results based on the calculated prevalence index of each pain point and the urgency index of each pain point, thereby obtaining the pain point identification results of the target product.
[0067] The product pain point identification method provided in the embodiment of the present application matches the target product review text with negative emotions with pain point through the pain point tag library, and then grades the pain points based on the universal index and the urgency index, thereby obtaining the target product pain point identification result, which can improve the accuracy of product pain point identification, improve the readability of pain points, and avoid the high cost of manual interpretation and correction. In addition, by introducing universality and urgency to grade pain points and quantify the value of pain points, it is helpful to discover new groups, new scenarios, etc., expand new product opportunities, improve the efficiency of decision-making on pain point processing, and significantly guide the optimization of enterprise pain point resource allocation and improve the competitiveness of enterprise products.
[0068] In one embodiment, based on the universality index and the urgency index of each pain point, the pain point matching results are graded to obtain the target product pain point identification results, including:
[0069] Based on the universality index of each pain point, determine the average universality index of all pain points in the pain point matching results; and based on the urgency index of each pain point, determine the average urgency index of all pain points in the pain point matching results;
[0070] Based on the average of the universality index of all pain points and the average of the urgency index of all pain points, determine the secondary pain point classification map;
[0071] Based on the positions of the universal index and the urgency index of each pain point in the secondary pain point classification map, the pain point matching results are graded to obtain the pain point identification results of the target product; the pain point identification results include core pain points, popular pain points, urgent pain points and marginal pain points; core pain points refer to pain points whose universal index is greater than or equal to the average of the universal indexes of all pain points, and whose urgency index is greater than the average of the urgency indexes of all pain points; popular pain points refer to pain points whose universal index is greater than or equal to the average of the universal indexes of all pain points, and whose urgency index is less than or equal to the average of the urgency indexes of all pain points; urgent pain points refer to pain points whose universal index is less than the average of the universal indexes of all pain points, and whose urgency index is greater than the average of the urgency indexes of all pain points; marginal pain points refer to pain points whose universal index is less than the average of the universal indexes of all pain points, and whose urgency index is less than or equal to the average of the urgency indexes of all pain points.
[0072] The pain point identification system can calculate the universality index of each pain point, find its average value, and use it as the average value of the universality index of all pain points in the pain point matching results; and can calculate the urgency index of each pain point, find its average value, and use it as the average value of the urgency index of all pain points in the pain point matching results. Then, based on the average value of the universality index of all pain points and the average value of the urgency index of all pain points, a secondary pain point classification map is determined. Specifically, the x-axis can represent the universality index, and the y-axis can represent the urgency index. The four quadrants can be divided according to the average value of the universality index of all pain points and the average value of the urgency index of all pain points to obtain a secondary pain point classification map (such as Figure 3 as shown).
[0073] The pain point identification system can determine the quadrant of the pain point based on the position of the universal index and the urgency index of each pain point in the secondary pain point classification map, and then classify the pain point matching results to obtain the target product pain point identification results. Based on the universal index and urgency index of each pain point, the pain point identification results can be specifically divided into core pain points, popular pain points, urgent pain points and marginal pain points (such as Figure 4 as shown).
[0074] Specifically, core pain points can refer to pain points that are prevalent and have an increasing negative impact on users, and can be used as the first pain points to be solved. Mass pain points can refer to pain points that are prevalent but have limited negative impact on users. Mass pain points are generally common problems in the industry and require planned and long-term gradual improvements to avoid them turning into core pain points due to poor handling. Emergency pain points can refer to pain points that only appear in a small number of scenarios, but once they appear, they have a large negative impact on consumers. Emergency pain points can be selectively solved, and if properly handled, they can become the focus of product differentiation. Marginal pain points can refer to pain points that only appear in a small number of scenarios and have limited negative impact on consumers. Generally, resources can be allocated to marginal pain points first, and the handling priority can be lowered, but they can be tracked.
[0075] The product pain point identification method provided in the embodiment of the present application determines the secondary pain point classification map through the mean of the universal index and the mean of the urgency index, thereby classifying the pain points and obtaining core pain points, popular pain points, emergency pain points and marginal pain points. By quantifying the value of pain points, it helps to discover new groups of people, new scenarios, etc., expand new product opportunities, further improve the efficiency of pain point processing decision-making, significantly guide the optimization of resource allocation for corporate pain points, and further improve the competitiveness of corporate products.
[0076] In one embodiment, based on the pain point tag library, the target product review text with negative emotions within a preset time period is matched with pain points to obtain the target product pain point matching results, including:
[0077] Based on the industry general tag library and industry characteristic tag library in the pain point tag library, pain point matching is performed on the target product review text with negative emotions within a preset time period to obtain the initial target product pain point matching result;
[0078] Based on the inapplicable tag library in the pain point tag library, the inapplicable pain point words in the initial target product pain point matching results are filtered to obtain the target product pain point matching results.
[0079] When performing pain point matching, the pain point identification system can first perform pain point matching on the target product review text with negative emotions within a preset time period based on the industry general tag library and the industry characteristic tag library in the pain point tag library, for example, using a naive algorithm, KMP (Knuth-Morris-Pratt Algorithm) algorithm, BM (Boyer-Moore Algorithm) algorithm, etc. to obtain the initial target product pain point matching result.
[0080] The initial target product pain point matching results may contain meaningless labels that match the industry's general label library. These labels may be mismatched due to the word segmentation relationship, resulting in confusion in the construction relationship. The pain point identification system can match the initial target product pain point matching results based on the inapplicable label library in the pain point label library, and filter the inapplicable pain point words that match the inapplicable label library in the initial target product pain point matching results, thereby obtaining the target product pain point matching results. For example: If the target product is a facial cleanser, the oily skin group matches the pain point word dryness, but dryness of facial cleanser is not a pain point for oily skin, so it can be filtered through the inapplicable label library.
[0081] The product pain point identification method provided in the embodiment of the present application matches product pain points through an industry general tag library, an industry characteristic tag library and an inapplicable tag library. It can not only match product pain points, but also filter out meaningless pain points, thereby further improving the accuracy of product pain point identification, improving the readability of pain points, and avoiding no-match problems and high-cost problems of manual interpretation and correction.
[0082] In one embodiment, based on the industry general tag library and the industry characteristic tag library in the pain point tag library, pain point matching is performed on the target product review text with negative sentiment within a preset time period, including:
[0083] Generate a first vector based on the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library; and generate a second vector based on the target keywords of the target product review text with negative emotions within a preset time period; the target keywords represent feedback on the target product;
[0084] A canonical match is performed on the first vector and the second vector.
[0085] The industry general tag library in the pain point tag library can contain general product pain point words, and the industry characteristic tag library can contain characteristic product pain point words. When matching the product pain point words corresponding to the industry general tag library and the industry characteristic tag library with the target keywords of the target product review text with negative emotions within a preset time period, the product pain point words and the target keywords can be vectorized first, and then regular matching can be performed.
[0086] The pain point identification system can generate a first vector based on the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library; and generate a second vector based on the target keyword of the target product review text with negative emotions in a preset time period, and then perform regular matching on the first vector and the second vector to obtain the initial target product pain point matching result. Among them, the target keyword can represent feedback on the target product. For example: through regular matching, "inappropriate" can be matched with "not very suitable".
[0087] The product pain point identification method provided in the embodiment of the present application can obtain the initial target product pain point matching result by vectorizing and regularizing the product pain point words and the target keywords, thereby further improving the accuracy of product pain point identification.
[0088] In one embodiment, the product pain point identification method further includes:
[0089] Determine that the similarity between the target keyword and any product pain point word is greater than a similarity threshold, and map the target keyword to the corresponding product pain point word.
[0090] In some cases, the following phenomenon may exist: the target keyword of the target product review text with negative emotions within a preset time period is inconsistent with the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library, but the target keyword and the product pain point word have the same or similar meaning. Matching the target keyword with the product pain point word will make it impossible to count the target keyword in the product pain point matching results, affecting the accuracy of product pain point identification.
[0091] Based on this, the product pain point identification method provided in the embodiment of the present application can first calculate the similarity between the target keyword and the product pain point word. When the similarity between the target keyword and a product pain point word is greater than the similarity threshold, the target keyword can be mapped to the corresponding product pain point word.
[0092] For example, vocabulary similarity calculations can be performed based on semantic dictionaries, corpus statistics, the number of search pages, etc. The similarity threshold can be set according to the specific situation, and this application does not make specific limitations. When the pain point identification system calculates that the similarity between the target keyword and a product pain point word is greater than the similarity threshold, the target keyword can be mapped to the corresponding product pain point word. For example, for a certain brand of lipstick, the product pain point word stored in the pain point identification system is "uncomfortable", and the similar target keyword is "not silky". The similarity between "uncomfortable" and "not silky" is greater than the threshold, then the pain point identification system can map the target keyword "not silky" to the corresponding product pain point word "uncomfortable".
[0093] The implementation of the technical solution provided in the embodiment of the present application can also be based on the following method: directly matching the product pain point words with the target keywords. For the target keywords that are not matched with the product pain point words, the target keywords can be further matched with the product pain point words for similarity. When the similarity between the target keyword and a product pain point word is greater than a preset similarity threshold, the target keyword can be mapped to the product pain point word.
[0094] The pain point identification system can first match the stored product pain point words with the target keywords. Due to the vocabulary of the product pain point words stored in the pain point identification system, there may be a situation where the target keywords and product pain point words are different and cannot be matched. The target keywords that cannot be matched may be words with the same or similar meanings as the product pain point words. For target keywords that cannot be matched, the pain point identification system can perform similarity matching on them. For example, for a certain brand of shower gel, the product pain point word is "bad smell" and the target keyword is "strange smell". Since the two words are different, when the pain point identification system directly matches the two, the target keyword "strange smell" and the product pain point word "bad smell" cannot be matched. At this time, the pain point identification system can further perform similarity matching on the target keyword "strange smell" and the product pain point word "bad smell". It is concluded that the similarity between "strange smell" and "bad smell" is greater than the similarity threshold. Then the pain point identification system can map the target keyword "strange smell" to the corresponding product pain point word "bad smell".
[0095] The product pain point identification method provided in the embodiment of the present application can further expand the scope of pain point identification by matching and mapping the similarity between the target keywords and the product pain point words, effectively avoiding the situation where the pain points of the review text are not identified due to limited product pain point words, thereby improving the accuracy of product pain point identification.
[0096] In one embodiment, the product pain point identification method further includes:
[0097] Determine that the ratio of the number of all pain points in the target product pain point matching results to the number of target product review texts with negative emotions within a preset time period is less than a preset ratio, and based on the target keywords in the target product review texts with negative emotions within the preset time period, update the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library.
[0098] When the ratio of the total number of pain points in the target product pain point matching results to the number of target product review texts with negative emotions within a preset time period is less than a preset ratio, the product pain point words corresponding to the target product can be updated based on the target keywords.
[0099] Among them, the preset ratio can be specifically 40%, 50%, etc., and its size can be adjusted according to actual conditions. This application does not make any specific limitations on this.
[0100] After the pain point identification system obtains the target product pain point matching results, it can calculate the ratio of the number of all pain points in the target product pain point matching results to the number of target product review texts with negative emotions within a preset time period. When the ratio of the number of all pain points to the number of target product review texts with negative emotions is less than a certain value, there may be a situation where the product pain point words are not rich enough. At this time, the product pain point words corresponding to the target product can be updated based on the target keywords.
[0101] When updating product pain point words based on target keywords, you can refer to certain standards, such as the frequency or quantity of target keywords. When the frequency of target keywords is high or the quantity is large, you can use it to update product pain point words.
[0102] For example, if the target keyword is "strange smell", "strange smell" is a word that describes the pain point of the product and its frequency of appearance is relatively high (for example, its frequency of appearance reaches 10% among all target keywords, or its frequency of appearance is relatively high in a certain time period: for example, the comment text obtained by the pain point identification system is from February to June, and "strange smell" appears more frequently in June); or there are many "strange smell" target keywords (for example, among 100 target keywords, 15 target keywords are "strange smell"), then "strange smell" can be updated to the industry general tag library and industry characteristic tag library in the pain point tag library.
[0103] The product pain point identification method provided in the embodiment of the present application can effectively improve the robustness of the product pain point identification method and continuously improve the accuracy of product pain point identification by continuously updating and improving product pain point words.
[0104] In one embodiment, the product pain point identification method further includes:
[0105] Based on the negative sentiment words of the industry to which the target product belongs, negative sentiment words are matched on the target product review texts within a preset time period to determine the target product review texts with negative sentiment.
[0106] Target product review texts with negative sentiment are filtered out from target product review texts within a preset time period. Negative sentiment words can be matched on the target product review texts based on negative sentiment words of the industry to which the target product belongs. Review texts including negative sentiment words are target product review texts with negative sentiment.
[0107] Before filtering out target product review texts with negative sentiment from the target product review texts, it is also possible to first obtain review text samples of the target products within a preset time period and preprocess the review text samples. The preprocessed review text samples can be used as the target product review texts.
[0108] The pain point identification system can extract the review text of the target product within a preset time period as a sample, and then perform pre-processing steps such as data deduplication, symbol and emoticon processing, meaningless text processing, and text data normalization processing on the sample data. Meaningless text processing includes processing of default evaluations, etc. The pre-processed review text sample is the review text of the target product.
[0109] The pain point identification system can extract and store negative sentiment words of the industry to which the target product belongs, and match them with the target product review text for negative sentiment words. If they match, the review text including negative sentiment words can be used as the target product review text with negative sentiment. If they do not match, the review text of the target product has positive sentiment, and it does not contain product pain points. For example, if the review text of the target product is "My skin is dry and it is OK to use it", the prior art usually recognizes that it contains product pain points because of the word "dry", while the present application first matches the review text with negative sentiment words, and recognizes that the review text cannot be matched with negative sentiment words, and the text sentiment is positive, then its recognition result is that it does not contain product pain points.
[0110] The product pain point identification method provided in the embodiment of the present application avoids the pain point identification of the review text with positive emotions by first identifying the negative emotions of the target product review text and then performing pain point matching, thereby improving the efficiency and accuracy of product pain point identification.
[0111] The following is an example of a product pain point identification method provided by an embodiment of the present application to further illustrate the technical solution of the present application.
[0112] like Figure 5 As shown, the example may include the following steps:
[0113] (1) Establish a pain point label library: Split the scene pain points according to the level of population needs, associate them with specific entities, and clearly display the corresponding pain points, as shown in the following table:
[0114]
[0115] Based on the pain point segmentation results, build a pain point label library.
[0116] (2) Text data input: Extract public opinion data of a specific industry within the processing time range, and perform data deduplication, symbol and emoticon processing, meaningless text processing (such as default evaluation, etc.), and text data normalization to obtain comment text.
[0117] (3) Text sentiment judgment: extract negative words from the sentiment vocabulary of the industry to which the target product belongs and concatenate them into a text array. Use the text array to perform vectorized matching on the comment text, and return a list. The length of the list corresponds to the number of negative words matched in the comment text. The value of each element in the list is the number of negative words matched in the comment text. Therefore, a comment text with a list element value greater than 0 is a comment text with negative sentiment.
[0118] (4) Output negative sentiment text: Extract the element numbers of the list whose element values are greater than 0 to form a vector m. Use vector m to index the comment text, and the first comment text df with negative sentiment is extracted.
[0119] (5) Pain point matching: Structural construction is performed in the form of entity-concept-relationship, product pain point words are extracted and concatenated into a text array, and the target keyword of the first comment text is vectorized and matched using the pain point tag library. The result is returned as a list. The length of the list corresponds to the number of pain point phrases matched by the target keyword of the first comment text. The value of each element in the list is the number of pain point words matched for each product.
[0120] (6) Output matching phrases: Extract the element numbers of the result list whose number of elements is greater than 0 to form a vector n. Use the vector n to index the first comment text df to obtain the text data res. Concatenate the text data res and the corresponding list elements to obtain a two-dimensional data table containing the text content and the matched pain point phrases.
[0121] (7) Matching the mapping relationship: Perform similarity matching between the target keywords and pain point phrases that are not matched with the product pain point words in the first comment text. When the similarity between the two is greater than the threshold, the product pain point results are modified based on the number of target keywords.
[0122] (8) Output product pain points: filter inapplicable tags and output product pain point matching results.
[0123] (9) Product pain point classification: Calculate the prevalence index and urgency index for the preset time period, construct a four-quadrant diagram for pain point classification, divide pain points into marginal pain points, popular pain points, urgent pain points and core pain points, and provide pain point classification and treatment suggestions.
[0124] The product pain point identification device provided in an embodiment of the present application is described below. The product pain point identification device described below and the product pain point identification method described above can be referenced to each other.
[0125] Figure 6 Schematic diagram of the structure of the product pain point identification device provided in the embodiment of the present application. Figure 6 , the device may include:
[0126] The matching module 610 is used to perform pain point matching on the target product review text with negative emotions within a preset time period based on the pain point tag library to obtain the target product pain point matching result; the pain point tag library includes an industry general tag library, an industry characteristic tag library and an inapplicable tag library; the industry general tag library is a tag library determined based on the general pain point words of the industry to which the target product belongs; the industry characteristic tag library is a tag library determined based on the characteristic pain point words of the industry to which the target product belongs; the inapplicable tag library is a tag library determined based on the inapplicable pain point words of the target product in the general pain point words of the industry general tag library;
[0127] The grading module 620 is used to grade the pain points of the pain point matching results based on the prevalence index of each pain point and the urgency index of each pain point to obtain the target product pain point identification result; the prevalence index of each pain point is determined based on the ratio of the number of each pain point in the pain point matching results to the total number of pain points; the urgency index of each pain point is determined based on the growth rate of the number of each pain point in the pain point matching results compared to the corresponding number of pain points in the previous preset time period.
[0128] The product pain point identification device provided in the embodiment of the present application matches the target product review text with negative emotions with pain point through the pain point tag library, and then grades the pain points based on the universal index and the urgency index, thereby obtaining the target product pain point identification result, which can improve the accuracy of product pain point identification, improve the readability of pain points, and avoid the high cost of manual interpretation and correction. In addition, by introducing universality and urgency to grade pain points and quantify the value of pain points, it is helpful to discover new groups, new scenarios, etc., expand new product opportunities, improve the efficiency of pain point processing decision-making, and significantly guide enterprises to optimize resource allocation based on pain points and improve the competitiveness of enterprise products.
[0129] In one embodiment, the classification module 620 is specifically used to:
[0130] Based on the universality index of each pain point, determine the average universality index of all pain points in the pain point matching result; and based on the urgency index of each pain point, determine the average urgency index of all pain points in the pain point matching result;
[0131] Determine a secondary pain point classification map based on the average of the universality index of all the pain points and the average of the urgency index of all the pain points;
[0132] Based on the positions of the prevalence index of each pain point and the urgency index of each pain point in the secondary pain point classification map, the pain point matching results are graded to obtain the target product pain point identification results; the pain point identification results include core pain points, popular pain points, urgent pain points and marginal pain points; the core pain points represent pain points whose prevalence index is greater than or equal to the average of the prevalence indexes of all pain points, and whose urgency index is greater than the average of the urgency indexes of all pain points; the popular pain points represent pain points whose prevalence index is greater than or equal to the average of the prevalence indexes of all pain points, and whose urgency index is less than or equal to the average of the urgency indexes of all pain points; the urgent pain points represent pain points whose prevalence index is less than the average of the prevalence indexes of all pain points, and whose urgency index is greater than the average of the urgency indexes of all pain points; the marginal pain points represent pain points whose prevalence index is less than the average of the prevalence indexes of all pain points, and whose urgency index is less than or equal to the average of the urgency indexes of all pain points.
[0133] In one embodiment, the matching module 610 is specifically used for:
[0134] Based on the industry general tag library and industry characteristic tag library in the pain point tag library, pain point matching is performed on the target product review text with negative emotions within a preset time period to obtain the initial target product pain point matching result;
[0135] Based on the inapplicable tag library in the pain point tag library, the inapplicable pain point words in the initial target product pain point matching result are filtered to obtain the target product pain point matching result.
[0136] In one embodiment, the matching module 610 is further configured to:
[0137] Generate a first vector based on the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library; and generate a second vector based on the target keywords of the target product review text with negative emotions within a preset time period; the target keywords represent feedback on the target product;
[0138] Perform canonical matching on the first vector and the second vector.
[0139] In one embodiment, the matching module 610 is further configured to:
[0140] It is determined that the similarity between the target keyword and any product pain point word is greater than a similarity threshold, and the target keyword is mapped to the corresponding product pain point word.
[0141] In one embodiment, the matching module 610 is further configured to:
[0142] Determine that the ratio of the total number of pain points in the target product pain point matching results to the number of target product review texts with negative emotions within a preset time period is less than a preset ratio, and based on the target keywords in the target product review texts with negative emotions within the preset time period, update the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library.
[0143] In one embodiment, the matching module 610 is further configured to:
[0144] Based on the negative sentiment words of the industry to which the target product belongs, negative sentiment words are matched on the target product review texts within a preset time period to determine the target product review texts with negative sentiment.
[0145] It should be noted here that the product pain point identification device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as those in the method embodiment will not be described in detail here.
[0146] Figure 7 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communication interface 720 and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute the following method:
[0147] Based on the pain point tag library, the target product review text with negative emotions within a preset time period is matched with pain points to obtain the target product pain point matching result; the pain point tag library includes an industry general tag library, an industry characteristic tag library and an inapplicable tag library; the industry general tag library is a tag library determined based on the general pain point words of the industry to which the target product belongs; the industry characteristic tag library is a tag library determined based on the characteristic pain point words of the industry to which the target product belongs; the inapplicable tag library is a tag library determined based on the inapplicable pain point words of the target product in the general pain point words of the industry general tag library;
[0148] Based on the universality index and the urgency index of each pain point, the pain point matching results are graded to obtain the pain point identification results of the target product; the universality index of each pain point is determined based on the ratio of the number of each pain point in the pain point matching results to the total number of pain points; the urgency index of each pain point is determined based on the growth rate of the number of each pain point in the pain point matching results compared to the corresponding number of pain points in the previous preset time period.
[0149] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the relevant technology. The computer software product is stored in a storage medium, including several instructions to enable 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 method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.
[0150] On the other hand, an embodiment of the present application discloses a computer program product, the computer program product including a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, when the program instructions are executed by a computer, the computer can perform the methods provided by the above-mentioned method embodiments, for example, including:
[0151] Based on the pain point tag library, the target product review text with negative emotions within a preset time period is matched with pain points to obtain the target product pain point matching result; the pain point tag library includes an industry general tag library, an industry characteristic tag library and an inapplicable tag library; the industry general tag library is a tag library determined based on the general pain point words of the industry to which the target product belongs; the industry characteristic tag library is a tag library determined based on the characteristic pain point words of the industry to which the target product belongs; the inapplicable tag library is a tag library determined based on the inapplicable pain point words of the target product in the general pain point words of the industry general tag library;
[0152] Based on the universality index and the urgency index of each pain point, the pain point matching results are graded to obtain the pain point identification results of the target product; the universality index of each pain point is determined based on the ratio of the number of each pain point in the pain point matching results to the total number of pain points; the urgency index of each pain point is determined based on the growth rate of the number of each pain point in the pain point matching results compared to the corresponding number of pain points in the previous preset time period.
[0153] In another aspect, an embodiment of the present application further provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method provided in each of the above embodiments is implemented, for example, including:
[0154] Based on the pain point tag library, the target product review text with negative emotions within a preset time period is matched with pain points to obtain the target product pain point matching result; the pain point tag library includes an industry general tag library, an industry characteristic tag library and an inapplicable tag library; the industry general tag library is a tag library determined based on the general pain point words of the industry to which the target product belongs; the industry characteristic tag library is a tag library determined based on the characteristic pain point words of the industry to which the target product belongs; the inapplicable tag library is a tag library determined based on the inapplicable pain point words of the target product in the general pain point words of the industry general tag library;
[0155] Based on the universality index and the urgency index of each pain point, the pain point matching results are graded to obtain the pain point identification results of the target product; the universality index of each pain point is determined based on the ratio of the number of each pain point in the pain point matching results to the total number of pain points; the urgency index of each pain point is determined based on the growth rate of the number of each pain point in the pain point matching results compared to the corresponding number of pain points in the previous preset time period.
[0156] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0157] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.
[0158] Finally, it should be noted that the above implementation modes are only used to illustrate the present application, rather than to limit the present application. Although the present application is described in detail with reference to the embodiments, a person skilled in the art should understand that various combinations, modifications or equivalent substitutions of the technical solutions of the present application do not depart from the spirit and scope of the technical solutions of the present application, and should be included in the scope of the claims of the present application.
Claims
1. A method for identifying product pain points, characterized in that: include: Based on the pain point tag library, the target product review text with negative emotions within a preset time period is matched with pain points to obtain the target product pain point matching results; The pain point tag library includes an industry general tag library, an industry characteristic tag library and an inapplicable tag library; the industry general tag library is a tag library determined based on the general pain point words of the industry to which the target product belongs; the industry characteristic tag library is a tag library determined based on the characteristic pain point words of the industry to which the target product belongs; the inapplicable tag library is a tag library determined based on the inapplicable pain point words of the target product in the general pain point words of the industry general tag library; Based on the prevalence index of each pain point and the urgency index of each pain point, the pain point matching results are graded to obtain the target product pain point identification results; the prevalence index of each pain point is determined based on the ratio of the number of each pain point in the pain point matching results to the number of all pain points; the urgency index of each pain point is determined based on the growth rate of the number of each pain point in the pain point matching results compared to the number of corresponding pain points in the previous preset time period; Based on the pain point tag library, the target product review text with negative emotions within a preset time period is subjected to pain point matching to obtain the target product pain point matching result, including: Based on the industry general tag library and the industry characteristic tag library in the pain point tag library, pain point matching is performed on the target product review text with negative emotions within a preset time period to obtain the initial target product pain point matching result; based on the inapplicable tag library in the pain point tag library, inapplicable pain point words in the initial target product pain point matching result are filtered to obtain the target product pain point matching result; The pain point matching is performed on the target product review text with negative emotions within a preset time period based on the industry general tag library and the industry characteristic tag library in the pain point tag library, including: Generate a first vector based on the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library; and generate a second vector based on the target keywords of the target product review text with negative emotions within a preset time period; the target keywords represent feedback on the target product; Performing canonical matching on the first vector and the second vector; The method further comprises: Determine that the similarity between the target keyword and any product pain point word is greater than a similarity threshold, and map the target keyword to the corresponding product pain point word; The method further comprises: Determine that the ratio of the total number of pain points in the target product pain point matching results to the number of target product review texts with negative emotions within a preset time period is less than a preset ratio, and based on the target keywords in the target product review texts with negative emotions within the preset time period, update the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library.
2. The product pain point identification method according to claim 1, characterized in that: The pain point matching results are graded based on the universality index and the urgency index of each pain point to obtain the target product pain point identification results, including: Based on the universality index of each pain point, determine the average universality index of all pain points in the pain point matching result; and based on the urgency index of each pain point, determine the average urgency index of all pain points in the pain point matching result; Determine a secondary pain point classification map based on the average of the universality index of all the pain points and the average of the urgency index of all the pain points; Based on the positions of the prevalence index of each pain point and the urgency index of each pain point in the secondary pain point classification map, the pain point matching results are graded to obtain the target product pain point identification results; the pain point identification results include core pain points, popular pain points, urgent pain points and marginal pain points; the core pain points represent pain points whose prevalence index is greater than or equal to the average of the prevalence indexes of all pain points, and whose urgency index is greater than the average of the urgency indexes of all pain points; the popular pain points represent pain points whose prevalence index is greater than or equal to the average of the prevalence indexes of all pain points, and whose urgency index is less than or equal to the average of the urgency indexes of all pain points; the urgent pain points represent pain points whose prevalence index is less than the average of the prevalence indexes of all pain points, and whose urgency index is greater than the average of the urgency indexes of all pain points; the marginal pain points represent pain points whose prevalence index is less than the average of the prevalence indexes of all pain points, and whose urgency index is less than or equal to the average of the urgency indexes of all pain points.
3. The product pain point identification method according to claim 1, characterized in that: The method further includes: performing negative sentiment word matching on target product review texts within a preset time period based on negative sentiment words of the industry to which the target product belongs, and determining target product review texts with negative sentiment.
4. A product pain point identification device, characterized in that: include: A matching module is used to perform pain point matching on target product review texts with negative emotions within a preset time period based on a pain point tag library to obtain target product pain point matching results; The pain point tag library includes an industry general tag library, an industry characteristic tag library and an inapplicable tag library; the industry general tag library is a tag library determined based on the general pain point words of the industry to which the target product belongs; the industry characteristic tag library is a tag library determined based on the characteristic pain point words of the industry to which the target product belongs; the inapplicable tag library is a tag library determined based on the inapplicable pain point words of the target product in the general pain point words of the industry general tag library; A grading module, configured to grade the pain points of the pain point matching results based on the prevalence index of each pain point and the urgency index of each pain point, so as to obtain the target product pain point identification result; the prevalence index of each pain point is determined based on the ratio of the number of each pain point in the pain point matching results to the number of all pain points; the urgency index of each pain point is determined based on the growth rate of the number of each pain point in the pain point matching results compared with the number of corresponding pain points in the previous preset time period; The matching module is specifically used for: Based on the industry general tag library and industry characteristic tag library in the pain point tag library, pain point matching is performed on the target product review text with negative emotions within a preset time period to obtain the initial target product pain point matching result; Based on the inapplicable tag library in the pain point tag library, filtering the inapplicable pain point words in the initial target product pain point matching result to obtain the target product pain point matching result; The matching module is specifically used for: Generate a first vector based on the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library; and generate a second vector based on the target keywords of the target product review text with negative emotions within a preset time period; the target keywords represent feedback on the target product; Performing canonical matching on the first vector and the second vector; The matching module is also used for: Determine that the similarity between the target keyword and any product pain point word is greater than a similarity threshold, and map the target keyword to the corresponding product pain point word; The matching module is also used for: Determine that the ratio of the total number of pain points in the target product pain point matching results to the number of target product review texts with negative emotions within a preset time period is less than a preset ratio, and based on the target keywords in the target product review texts with negative emotions within the preset time period, update the product pain point words corresponding to the industry general tag library and the industry characteristic tag library in the pain point tag library.
5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the product pain point identification method according to any one of claims 1 to 3 is implemented.
6. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the product pain point identification method according to any one of claims 1 to 3 is implemented.
