Intelligent recommendation method, system and storage medium for industrial design products

Abstract

The application provides an intelligent recommendation method and system for an industrial design product and a storage medium, relates to the technical field of product recommendation, and comprises the following steps: acquiring voice data of a user and comprehensively analyzing the voice data to obtain a keyword selection coefficient; selecting a keyword by threshold comparison of the keyword selection coefficient, taking the selected keyword as an index to search an industrial design product library to obtain a first recommended product set; analyzing matching between a user's expected price and material and product price and material to obtain a second recommended product set; analyzing user interaction data to obtain an interaction score; and recommending an industrial design product with the largest interaction score in the second recommended product set to the user. The application improves the relevance of the recommended result and user satisfaction by in-depth demand analysis and research on the matching between demand and the recommended result.

Description

Technical Field

[0001] This invention relates to the field of product recommendation technology, specifically to intelligent recommendation methods, systems, and storage media for industrial design products. Background Technology

[0002] With the increasing variety of industrial design products, users face the challenge of navigating a vast amount of information and complex filtering when choosing products. Traditional recommendation methods often rely on collaborative filtering or content matching techniques based on user behavior. However, these methods frequently fail to accurately capture users' true needs, especially when user intent is ambiguous, resulting in low accuracy and user satisfaction. In recent years, the development of voice interaction technology has provided a more natural way for users to interact with systems, making voice-based product recommendation an important research direction for improving user experience. However, current voice recommendation methods mostly remain at the level of simple keyword matching, failing to fully explore the deeper relationship between voice features and user needs, thus the accuracy of recommendation results still needs improvement.

[0003] In the prior art, CN116738864B discloses an intelligent recommendation method and system for industrial design products. This method includes the following steps: acquiring industrial design demand data and performing deep analysis on the data to obtain a semantic graph of design demands; matching material properties with design constraints based on the semantic graph to obtain a constraint-compliant material list; performing optimization algorithm derivation on the constraint-compliant material list to obtain an optimized design decision tree; collecting real-time production, usage environment, and user feedback data through IoT technology and integrating it with the optimized design decision tree to obtain a real-time design feedback database; and optimizing the design decisions based on the real-time design feedback database to obtain optimized design decision data. The prior art only performs a superficial analysis of semantic demands, failing to deeply mine the emotional tendencies within the semantics. This approach ignores the potential emotions and preferences of users when expressing their needs, resulting in inaccurate capture of users' true needs. Consequently, there is a significant deviation between the generated recommendation strategy and the user's actual intent, affecting the relevance of the recommendation results and the user experience. Furthermore, the prior art only focuses on the degree of material matching during the matching process, neglecting the crucial factor of price matching. Price is an indispensable part of the user's decision-making process, and its matching directly affects the product's cost-effectiveness and the user's willingness to buy. If price matching is ignored, the system may recommend products that meet the material requirements but deviate from the user's expectations in terms of price, thereby further reducing the overall accuracy and usability of the recommendation system.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent recommendation method and system for industrial design products to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A smart recommendation method for industrial design products, comprising the following steps:

[0008] Step 1: Build an industrial design product library and keyword library through the industry design library, obtain the user's voice data, and convert the user's voice data into text data; count the frequency of each keyword in the text data, and extract the corresponding voice part of each keyword each time it appears to obtain keyword voice data;

[0009] Step 2: Extract features from the keyword speech data to obtain keyword speech feature data. Perform comprehensive analysis on the keyword speech feature data and the frequency of each keyword to obtain the keyword selection coefficient. The keyword speech feature data includes the cepstral coefficients of the first 13 Mel frequencies and short-time energy.

[0010] Step 3: Select keywords by comparing the keyword selection coefficients with thresholds, and use the selected keywords as indexes to search the industrial design product database to obtain the first recommended product set;

[0011] Step 4: Obtain the user's ideal price and material data, compare the user's ideal price with the prices of each product in the first recommended product set to generate a price matching degree; compare the user's ideal material with the materials of each product in the first recommended product set to generate a material matching degree; generate a product matching degree based on the price matching degree and material matching degree; preset a product matching threshold; select industrial design products with a product matching degree greater than the product matching threshold from the first recommended product set to form the second recommended product set;

[0012] Step 5: Obtain the interaction data for each product in the second recommended product set, analyze the interaction data to obtain the interaction score for each product in the second recommended product set, and recommend the industrial design product with the highest interaction score in the second recommended product set to the user.

[0013] Furthermore, the industrial design product library includes industrial design products and their categories, functions, materials, prices, application scenarios, page views, collections, and purchases; the keyword library includes the categories, functions, materials, and application scenarios marked in the aforementioned industrial design product library.

[0014] Furthermore, the keyword speech data is divided into discrete keyword speech data by dividing it into 25ms time windows and sliding it with a fixed step size of 10ms. The short-time energy of the discrete keyword speech data is then calculated.

[0015] The specific formula used to calculate short-time energy is as follows:

[0016]

[0017] in, For the first The first keyword The short-term energy of the keyword's speech upon its first occurrence. The total number of discrete keyword speech data. For the first The first keyword When it appears for the first time, the first The decibel values ​​of discrete keyword speech data. An index for discrete keyword speech data. , For keyword index and , This represents the total number of keywords in the text data. An index for the frequency of keyword occurrences, and , This represents the total number of times the keyword appears.

[0018] The specific logic for generating the keyword selection coefficient is as follows: Calculate the average value of the first and second Mel frequency cepstral coefficients; the variance of the third to thirteenth Mel frequency cepstral coefficients; and generate the keyword selection coefficient based on the average value of the first and second Mel frequency cepstral coefficients, the variance of the third to thirteenth Mel frequency cepstral coefficients, the number of keyword occurrences, short-time energy, and the frequency of each keyword occurrence. The specific formula for generating the keyword selection coefficient is as follows:

[0019]

[0020] in, For the first Keyword selection coefficient for each keyword. For the first The frequency of each keyword For the first Total number of times each keyword appears For the first The first keyword The variance of the cepstral coefficients of the 3rd to 13th Mel frequencies in the speech of the keyword at the time of its occurrence. For the first The first keyword When the keyword appears for the first time, the average value of the cepstral coefficients of the first and second Mel-frequency components of the speech signal is given. For the first The first keyword The short-term energy of the keyword speech when it appears for the first time.

[0021] Furthermore, the specific logic for selecting keywords is as follows: a preset keyword selection threshold is set, the keyword selection coefficient is compared with the keyword selection threshold, and keywords with a keyword selection coefficient greater than the keyword selection threshold are selected keywords;

[0022] The specific process of searching the industrial design product database using selected keywords as indexes is as follows: Based on the magnitude of the keyword selection coefficient, the selected keywords are sorted in descending order. First, all selected keywords are entered into the search engine. If no results are found, the last selected keyword is removed, and the search continues. This process is repeated until search results appear.

[0023] Furthermore, the specific formula used to calculate the price matching degree is as follows:

[0024]

[0025]

[0026] in, The first recommended product is number one. Price matching of industrial design products The first recommended product is number one. The price of an industrial design product. For the user's ideal price, As a price correction factor, To match price sensitivity, , This is the primary recommended product collection, an index of industrial design products, and , The total number of industrial design products in the top recommended product category;

[0027] The specific formula used to calculate material matching degree is as follows:

[0028]

[0029] in, The first recommended product is number one. Material matching of industrial design products For users' ideal material set, The first recommended product is number one. A collection of materials used in industrial design products. For users' ideal material set and first recommended product set The quantity of materials and elements in an industrial design product For users' ideal material set and first recommended product set The number of elements in the material set of an industrial design product;

[0030] The specific formula used to generate product matching degree is as follows:

[0031]

[0032] in, The first recommended product is number one. Product matching degree of industrial design products.

[0033] Furthermore, the interaction data includes the number of views, favorites, and purchases for each industrial design product;

[0034] The specific formula used to calculate the interaction rating is as follows:

[0035]

[0036] in, For the second recommended product focus, the first Interactive ratings for industrial design products For the second recommended product focus, the first Page views of industrial design products For the second recommended product focus, the first The number of industrial design products collected. For the second recommended product focus, the first Purchase volume of industrial design products To Perform a floor operation. This is the second recommended product collection, an index of industrial design products, and , The second recommended product category is the total number of industrial design products.

[0037] This invention also provides an intelligent recommendation system for industrial design products. The system is used to implement the intelligent recommendation method for industrial design products, specifically including:

[0038] The data acquisition module is used to build an industrial design product library and a keyword library through the industry design library, acquire user voice data, convert user voice data into text data, count the frequency of each keyword in the text data, and extract the corresponding voice part of each keyword each time it appears to obtain keyword voice data.

[0039] The data analysis module is used to extract features from keyword speech data to obtain keyword speech feature data, and to perform comprehensive analysis on the keyword speech feature data and the frequency of each keyword to obtain the keyword selection coefficient; the keyword speech feature data includes the cepstral coefficients of the first 13 Mel frequencies and short-time energy;

[0040] The product search module is used to select keywords by comparing the keyword selection coefficient with a threshold. The selected keywords are used as indexes to search the industrial design product database to obtain the first recommended product set.

[0041] The matching analysis module is used to obtain the user's ideal price and material data, compare the user's ideal price with the prices of each product in the first recommended product set to generate a price matching degree; compare the user's ideal material with the materials of each product in the first recommended product set to generate a material matching degree; generate a product matching degree based on the price matching degree and the material matching degree; preset a product matching threshold; select industrial design products with a product matching degree greater than the product matching threshold from the first recommended product set to form a second recommended product set;

[0042] The interaction analysis module is used to acquire interaction data for each product in the second recommended product set, analyze the interaction data to obtain the interaction score for each product in the second recommended product set, and recommend the industrial design product with the highest interaction score in the second recommended product set to the user.

[0043] The present invention further provides a computer-readable storage medium, wherein the storage medium internally stores a computer program that can be executed by a processor, and the computer program, when executed by the processor, can implement the intelligent recommendation method for industrial design products.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] This invention incorporates user sentiment analysis into the keyword selection coefficient generation process during user demand analysis. This allows the system to capture the emotional inclinations within user semantics, thereby more accurately identifying users' true needs. Furthermore, the recommendation strategy not only focuses on material matching but also considers price factors, ensuring that recommendations satisfy both user preferences for materials and their price expectations. This comprehensive approach, considering both sentiment and price factors, significantly improves the recommendation system's ability to accurately analyze and match user needs, thereby enhancing the relevance of recommendations and user satisfaction. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the overall method flow of the present invention.

[0047] Figure 2 This is a schematic diagram of the overall system structure of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0049] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0050] Example:

[0051] Please see Figure 1 The present invention provides a technical solution:

[0052] A smart recommendation method for industrial design products, comprising the following steps:

[0053] Step 1: Build an industrial design product library and keyword library through the industry design library, obtain the user's voice data, and convert the user's voice data into text data; count the frequency of each keyword in the text data, and extract the corresponding voice part of each keyword each time it appears to obtain keyword voice data;

[0054] The industrial design product library includes industrial design products and their categories, functions, materials, prices, application scenarios, page views, collections, and purchases. The keyword library includes the categories, functions, materials, and application scenarios marked in the aforementioned industrial design product library. Step 2: Extract features from the keyword speech data to obtain keyword speech feature data. Perform a comprehensive analysis of the keyword speech feature data and the frequency of each keyword to obtain the keyword selection coefficient. The keyword speech feature data includes the cepstral coefficients of the first 13 Mel-frequency components and short-time energy.

[0055] The keyword speech data is divided into discrete segments by a 25ms time window and then slid in a fixed step of 10ms. The short-time energy of the discrete keyword speech data is then calculated.

[0056] The specific formula used to calculate short-time energy is as follows:

[0057]

[0058] in, For the first The first keyword The short-term energy of the keyword's speech upon its first occurrence. The total number of discrete keyword speech data. For the first The first keyword When it appears for the first time, the first The decibel values ​​of discrete keyword speech data. An index for discrete keyword speech data. , For keyword index and , This represents the total number of keywords in the text data. An index for the frequency of keyword occurrences, and , This represents the total number of times the keyword appears.

[0059] The method for obtaining the first 13 Mel-frequency cepstral coefficients is as follows: A short-time Fourier transform is performed on each segment of the speech signal to obtain the speech spectrum information. The spectrum is mapped to a Mel scale, the logarithm of the Mel spectrum is taken, a discrete cosine transform is calculated, the energy of the Mel spectrum is compressed, and a set of Mel-frequency cepstral coefficients is generated. These Mel-frequency cepstral coefficients are then arranged in descending order of energy, and the first 13 Mel-frequency cepstral coefficients are selected. These are all conventional techniques in this field and will not be elaborated upon here.

[0060] The specific logic for generating the keyword selection coefficient is as follows: Calculate the average value of the first and second Mel frequency cepstral coefficients; the variance of the third to thirteenth Mel frequency cepstral coefficients; and generate the keyword selection coefficient based on the average value of the first and second Mel frequency cepstral coefficients, the variance of the third to thirteenth Mel frequency cepstral coefficients, the number of keyword occurrences, short-time energy, and the frequency of each keyword occurrence. The specific formula for generating the keyword selection coefficient is as follows:

[0061]

[0062] in, For the first Keyword selection coefficient for each keyword. For the first The frequency of each keyword For the first Total number of times each keyword appears For the first The first keyword The variance of the cepstral coefficients of the 3rd to 13th Mel frequencies in the speech of the keyword at the time of its occurrence. For the first The first keyword When the keyword appears for the first time, the average value of the cepstral coefficients of the first and second Mel-frequency components of the speech signal is given. For the first The first keyword The short-term energy of the keyword speech when it appears for the first time.

[0063] Keyword selection coefficient reflects the importance or selection value of keywords in user voice input. The value of the keyword directly affects the relevance of subsequent product recommendations; the higher the value, the more important the keyword is to the user. The generation of this coefficient can provide an important basis for recommending industrial design products. The frequency of the keyword appearing in the user's speech-to-text conversion. High frequency indicates high user attention to the keyword and is a direct indicator of keyword importance in the recommendation system. The first two Mel-frequency cepstral coefficients reflect the overall energy distribution and approximate shape of the spectral envelope of the speech signal, including the pitch. Their average value reflects the stability of the speech pitch; the higher the value, the worse the pitch stability and the more intense the speech. The variance of the 3rd to 13th Mel-frequency cepstral coefficients reflects the energy changes in the high-frequency region of the speech. The mid-to-high frequency region is closely related to the timbre and emotional richness of the speech. A larger variance indicates a more complex timbre and richer emotional expression. The short-time energy of the keyword speech reflects the total energy of the keyword speech; the higher the value, the greater the total energy of the keyword speech and the greater the energy in the high-frequency region. The variance of the cepstral coefficients of the 3rd to 13th Mel frequencies is coupled with the short-time energy of the keyword speech to jointly describe the emotional expressiveness and complexity of the speech. The larger the value, the stronger the emotional expressiveness of the speech and the more complex it is. It reflects the degree of emotional richness; the higher the value, the richer the emotion in the voice.

[0064] Step 3: Select keywords by comparing the keyword selection coefficients with thresholds, and use the selected keywords as indexes to search the industrial design product database to obtain the first recommended product set;

[0065] The specific logic for selecting keywords is as follows: a keyword selection threshold is preset, the keyword selection coefficient is compared with the keyword selection threshold, and keywords with a keyword selection coefficient greater than the keyword selection threshold are selected keywords.

[0066] The specific process of searching the industrial design product database using selected keywords as indexes is as follows: Based on the magnitude of the keyword selection coefficient, the selected keywords are sorted in descending order. First, all selected keywords are entered into the search engine. If no results are found, the last selected keyword is removed, and the search continues. This process is repeated until search results appear.

[0067] Step 4: Obtain the user's ideal price and material data, compare the user's ideal price with the prices of each product in the first recommended product set to generate a price matching degree; compare the user's ideal material with the materials of each product in the first recommended product set to generate a material matching degree; generate a product matching degree based on the price matching degree and material matching degree; preset a product matching threshold; select industrial design products with a product matching degree greater than the product matching threshold from the first recommended product set to form the second recommended product set;

[0068] The specific formula used to calculate the price matching degree is as follows:

[0069]

[0070]

[0071] in, The first recommended product is number one. Price matching of industrial design products The first recommended product is number one. The price of an industrial design product. For the user's ideal price, As a price correction factor, To match price sensitivity, , This is the primary recommended product collection, an index of industrial design products, and , The total number of industrial design products in the top recommended product category;

[0072] Price fit indicates the degree to which a product's price matches a user's ideal price, ranging from 0 to 1. A value closer to 1 indicates a closer match between the product price and the user's ideal price; a value closer to 0 indicates a larger discrepancy. Price fit sensitivity reflects the sensitivity to price matching; it acts as a regulator, controlling the impact of price deviations on the fit. A smaller ε value makes the price fit more sensitive to deviations. Price is a crucial factor to consider when choosing a product; generally, the lower the price, the greater the probability that a user will choose the product. This is used to reflect the degree of deviation between the product price and the user's ideal price. However, this deviation can be high or low. If the product price is lower than the user's ideal price, the user is satisfied with the product price. Therefore, a price adjustment factor is used to maximize the price matching degree to 1, avoiding [the deviation being too large or too small]. It cannot reflect the degree of price deviation.

[0073] The specific formula used to calculate material matching degree is as follows:

[0074]

[0075] in, The first recommended product is number one. Material matching of industrial design products For users' ideal material set, The first recommended product is number one. A collection of materials used in industrial design products. For users' ideal material set and first recommended product set The quantity of materials and elements in an industrial design product For users' ideal material set and first recommended product set The number of elements in the material set of an industrial design product;

[0076] Material matching degree indicates the degree of matching between the user's ideal material and the product material, ranging from 0 to 1. The closer the value is to 1, the better the match between the user's ideal material and the product material; the closer the value is to 0, the lower the matching degree. This includes materials that exist simultaneously in both the user's ideal material set and the product material set. The more overlapping elements there are, the better. The higher the value, the better the material matching. This includes all unique materials present in the user's ideal material set or the product's material set. It is used to standardize the matching degree calculation results.

[0077] The specific formula used to generate product matching degree is as follows:

[0078]

[0079] in, The first recommended product is number one. Product matching degree of industrial design products.

[0080] Price matching degree indicates the degree of match between the product price and the user's ideal price, ranging from 0 to 1. The closer the value is to 1, the closer the product price is to the user's ideal price; the closer the value is to 0, the greater the gap. Material matching degree indicates the degree of match between the user's ideal material and the product material, ranging from 0 to 1. The closer the value is to 1, the better the match between the user's ideal material and the product material; the closer the value is to 0, the lower the match. Product matching degree combines price matching degree and material matching degree, indicating the overall degree of match between the product and the user's needs, ranging from 0 to 1. The closer the value is to 1, the better the match between the product and the user's preferences; the closer the value is to 0, the lower the match. The better the industrial design product matches the user's needs, the more suitable the product is to be recommended to the user. The generation of this coefficient can provide an important basis for the intelligent recommendation of industrial design products.

[0081] Step 5: Obtain the interaction data for each product in the second recommended product set, analyze the interaction data to obtain the interaction score for each product in the second recommended product set, and recommend the industrial design product with the highest interaction score in the second recommended product set to the user.

[0082] The interactive data includes the number of views, favorites, and purchases for each industrial design product.

[0083] The specific formula used to calculate the interaction rating is as follows:

[0084]

[0085] in, For the second recommended product focus, the first Interactive ratings for industrial design products For the second recommended product focus, the first Page views of industrial design products For the second recommended product focus, the first The number of industrial design products collected. For the second recommended product focus, the first Purchase volume of industrial design products To Perform a floor operation. This is the second recommended product collection, an index of industrial design products, and , The second recommended product category is the total number of industrial design products.

[0086] Interaction rating represents the overall interaction score of a product, reflecting the degree to which the product is popular with users. A higher number indicates that the product is popular among users and has a high number of views, favorites, and purchases. The views, favorites, and purchases refer to all views, favorites, and purchases of the industrial design product within the industrial design product database and among the searched products.

[0087] Please see Figure 2 The present invention further provides an intelligent recommendation system for industrial design products. This system is used to implement the intelligent recommendation method for industrial design products, specifically including:

[0088] The data acquisition module is used to build an industrial design product library and a keyword library through the industry design library, acquire user voice data, convert user voice data into text data, count the frequency of each keyword in the text data, and extract the corresponding voice part of each keyword each time it appears to obtain keyword voice data.

[0089] The data analysis module is used to extract features from keyword speech data to obtain keyword speech feature data, and to perform comprehensive analysis on the keyword speech feature data and the frequency of each keyword to obtain the keyword selection coefficient; the keyword speech feature data includes the cepstral coefficients of the first 13 Mel frequencies and short-time energy;

[0090] The product search module is used to select keywords by comparing the keyword selection coefficient with a threshold. The selected keywords are used as indexes to search the industrial design product database to obtain the first recommended product set.

[0091] The matching analysis module is used to obtain the user's ideal price and material data, compare the user's ideal price with the prices of each product in the first recommended product set to generate a price matching degree; compare the user's ideal material with the materials of each product in the first recommended product set to generate a material matching degree; generate a product matching degree based on the price matching degree and the material matching degree; preset a product matching threshold; select industrial design products with a product matching degree greater than the product matching threshold from the first recommended product set to form a second recommended product set;

[0092] The interaction analysis module is used to acquire interaction data for each product in the second recommended product set, analyze the interaction data to obtain the interaction score for each product in the second recommended product set, and recommend the industrial design product with the highest interaction score in the second recommended product set to the user.

[0093] The present invention further provides a computer-readable storage medium, wherein the storage medium internally stores a computer program that can be executed by a processor, and the computer program, when executed by the processor, can implement the intelligent recommendation method for industrial design products.

[0094] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0095] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0096] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0097] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that cannot be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for intelligent recommendation of industrial design products, characterized in that, The specific steps include: Step 1: Build an industrial design product library and keyword library through the industry design library, obtain the user's voice data, and convert the user's voice data into text data; count the frequency of each keyword in the text data, and extract the corresponding voice part of each keyword each time it appears to obtain keyword voice data; Step 2: Extract features from the keyword speech data to obtain keyword speech feature data. Perform comprehensive analysis on the keyword speech feature data and the frequency of each keyword to obtain the keyword selection coefficient. The keyword speech feature data includes the cepstral coefficients of the first 13 Mel frequencies and short-time energy. Step 3: Select keywords by comparing the keyword selection coefficients with thresholds, and use the selected keywords as indexes to search the industrial design product database to obtain the first recommended product set; Step 4: Obtain the user's ideal price and material data, compare the user's ideal price with the prices of each product in the first recommended product set to generate a price matching degree; compare the user's ideal material with the materials of each product in the first recommended product set to generate a material matching degree; generate a product matching degree based on the price matching degree and material matching degree; preset a product matching threshold; select industrial design products with a product matching degree greater than the product matching threshold from the first recommended product set to form the second recommended product set; Step 5: Obtain the interaction data of each product in the second recommended product set, analyze the interaction data to obtain the interaction score of each product in the second recommended product set, and recommend the industrial design product with the highest interaction score in the second recommended product set to the user; The keyword speech data is divided into discrete keyword speech data by dividing it into 25ms time windows and sliding it with a fixed step size of 10ms. The short-time energy of the discrete keyword speech data is then calculated. The specific formula used to calculate short-time energy is as follows: ； in, For the first The first keyword The short-term energy of the keyword's speech upon its first occurrence. The total number of discrete keyword speech data. For the first The first keyword When it appears for the second time, the first The decibel values ​​of discrete keyword speech data. An index for discrete keyword speech data. , For keyword index and , This represents the total number of keywords in the text data. This is an index based on the frequency of keyword occurrences, and , This represents the total number of times the keyword appears. The specific logic for generating the keyword selection coefficient is as follows: Calculate the average value of the first and second Mel frequency cepstral coefficients; the variance of the third to thirteenth Mel frequency cepstral coefficients; and generate the keyword selection coefficient based on the average value of the first and second Mel frequency cepstral coefficients, the variance of the third to thirteenth Mel frequency cepstral coefficients, the number of keyword occurrences, short-time energy, and the frequency of each keyword occurrence. The specific formula for generating the keyword selection coefficient is as follows: ； in, For the first Keyword selection coefficient for each keyword For the first The frequency of each keyword For the first Total number of times each keyword appears For the first The first keyword The variance of the cepstral coefficients of the 3rd to 13th Mel frequencies in the speech of the keyword at the time of its occurrence. For the first The first keyword When the keyword appears for the first time, the average value of the cepstral coefficients of the first and second Mel-frequency components of the speech signal is given. For the first The first keyword The short-term energy of the keyword speech when it first appears; The specific formula used to calculate the price matching degree is as follows: ； ； in, The first recommended product is number one. Price matching of industrial design products The first recommended product is number one. The price of an industrial design product. For the user's ideal price, As a price correction factor, To match price sensitivity, , This is the primary recommended product collection, an index of industrial design products, and , The total number of industrial design products in the top recommended product category; The specific formula used to calculate material matching degree is as follows: ； in, The first recommended product is number one. Material matching of industrial design products For users' ideal material set, The first recommended product is number one. A collection of materials used in industrial design products. For users' ideal material set and first recommended product set The quantity of materials and elements in an industrial design product For users' ideal material set and first recommended product set The number of elements in the material set of an industrial design product; The specific formula used to generate product matching degree is as follows: ； in, The first recommended product is number one. Product matching degree of industrial design products; The interactive data includes the number of views, favorites, and purchases for each industrial design product. The specific formula used to calculate the interaction rating is as follows: ； in, For the second recommended product focus, the first Interactive ratings for industrial design products For the second recommended product focus, the first Page views of industrial design products For the second recommended product focus, the first The number of industrial design products collected. For the second recommended product focus, the first Purchase volume of industrial design products To Perform a floor operation. This is the second recommended product collection, an index of industrial design products, and , The second recommended product category is the total number of industrial design products.

2. The intelligent recommendation method for industrial design products according to claim 1, characterized in that: The industrial design product library includes industrial design products and their categories, functions, materials, prices, application scenarios, page views, number of favorites, and number of purchases; the keyword library includes the categories, functions, materials, and application scenarios marked in the aforementioned industrial design product library.

3. The intelligent recommendation method for industrial design products according to claim 1, characterized in that: The specific logic for selecting keywords is as follows: a keyword selection threshold is preset, the keyword selection coefficient is compared with the keyword selection threshold, and keywords with a keyword selection coefficient greater than the keyword selection threshold are selected keywords. The specific process of searching the industrial design product database using selected keywords as indexes is as follows: Based on the magnitude of the keyword selection coefficient, the selected keywords are sorted in descending order. First, all selected keywords are entered into the search engine. If no results are found, the last selected keyword is removed, and the search continues. This process is repeated until search results appear.

4. An intelligent recommendation system for industrial design products, characterized in that: The system is used to implement the intelligent recommendation method for industrial design products as described in any one of claims 1-3, specifically including: The data acquisition module is used to build an industrial design product library and a keyword library through the industry design library, acquire user voice data, convert user voice data into text data, count the frequency of each keyword in the text data, and extract the corresponding voice part of each keyword each time it appears to obtain keyword voice data. The data analysis module is used to extract features from keyword speech data to obtain keyword speech feature data, and to perform comprehensive analysis on the keyword speech feature data and the frequency of each keyword to obtain the keyword selection coefficient; the keyword speech feature data includes the cepstral coefficients of the first 13 Mel frequencies and short-time energy; The product search module is used to select keywords by comparing the keyword selection coefficient with a threshold. The selected keywords are used as indexes to search the industrial design product database to obtain the first recommended product set. The matching analysis module is used to obtain the user's ideal price and material data, compare the user's ideal price with the prices of each product in the first recommended product set to generate a price matching degree; compare the user's ideal material with the materials of each product in the first recommended product set to generate a material matching degree; generate a product matching degree based on the price matching degree and the material matching degree; preset a product matching threshold; select industrial design products with a product matching degree greater than the product matching threshold from the first recommended product set to form a second recommended product set; The interaction analysis module is used to acquire interaction data for each product in the second recommended product set, analyze the interaction data to obtain the interaction score for each product in the second recommended product set, and recommend the industrial design product with the highest interaction score in the second recommended product set to the user.

5. A computer-readable storage medium, characterized in that: The storage medium internally stores a computer program that can be executed by a processor. When the computer program is executed by the processor, it can implement the intelligent recommendation method for industrial design products as described in any one of claims 1-3.

Images