CN118364115A - A Product Design Information Classification System - Google Patents

Abstract

The invention relates to the technical field of information classification, in particular to a product design information classification system, which comprises an information acquisition module, a functional classification module, a product feature analysis module, a user analysis module, a correlation analysis module and an information classification module, wherein the information acquisition module is used for acquiring product function information, product specification information and product identification information of a design product, the user information is also used for acquiring user information, the functional classification module is used for analyzing functional keywords of the design product and analyzing product types, the product feature analysis module is used for analyzing quality labels and performance labels of the design product, the product feature analysis module is also used for analyzing product features, the user analysis module is used for analyzing user selection states, the correlation analysis module is used for analyzing correlations between the design product and a user, the information classification module is used for carrying out matching classification on the design product, and the classification storage module is used for carrying out classification storage on product identification information of the design product. The invention effectively improves the classification efficiency of the product design information.

Description

A Product Design Information Classification System

Technical Field

The present invention relates to the technical field of information classification, and in particular to a product design information classification system.

Background technique

With the rapid development of the manufacturing industry, the product design process has become increasingly complex, involving a large amount of design documents, drawings, material specifications, cost budgets and other information. Existing information management methods often rely on manual classification and storage, which can easily lead to information confusion and difficulty in finding, seriously affecting the product development cycle and team collaboration efficiency. Therefore, there is an urgent need for a system that can automatically and intelligently classify and manage product design information.

Chinese patent publication number CN103927400A discloses a method for classifying and crawling detailed information of Web products and establishing a product information library. The invention designs a webpage crawling method for obtaining product information on Web websites. First, the homepage of the first-level classification of the website products is crawled, and the link to the next-level product classification homepage is obtained by analyzing the source file of the crawled classification homepage; then crawling step by step until all the classification homepages of the website are crawled; by analyzing the source files of all classification subpages, the page turning elements and the number of pages of each classification are obtained, and then the subpage links of each classification are generated, and finally, according to the subpage links of each classification, the crawling of each classification subpage is completed. At the same time, by analyzing the source files of the product classification subpages crawled by the crawler, the product detailed information and the classification information to which the product belongs are extracted, and the mapping relationship between the website product id, classification id and other detailed information is established to build a product information library; it can be seen that when classifying products, the invention does not analyze the correlation between users and products, and the model of product classification is single, which leads to the problem of low efficiency in classifying product design information.

Summary of the invention

To this end, the present invention provides a product design information classification system to overcome the problem of low efficiency in product design information classification in the prior art.

To achieve the above object, the present invention provides a product design information classification system, comprising:

An information acquisition module is used to acquire product function information, product specification information and product identification information of the designed product, and is also used to acquire user information;

A function classification module is used to analyze the function keywords of the design product according to the product function information of the design product, and to analyze the product category according to the function keyword analysis result of the design product;

A product feature analysis module, used to analyze the quality label and performance label of the designed product according to the product category analysis result and product specification information, and also used to analyze the product features according to the quality label and performance label analysis result of the designed product;

User analysis module, used to analyze user selection status based on user information;

A correlation analysis module is used to analyze the correlation between the designed product and the user based on the user selection status analysis results and the product feature analysis results;

An information classification module is used to match and classify design products according to the correlation analysis results between design products and users;

The classification storage module is used to classify and store the product identification information of the designed product according to the matching classification results of the designed product.

Further, the product feature analysis module is provided with a performance label analysis unit, and the performance label analysis unit analyzes the product performance label according to the product category analysis result and the product specification information;

The performance label analysis unit analyzes the performance label of the product according to the product category analysis result and the product specification information;

The performance label analysis unit calculates the product performance index α according to the product specification information. The calculation formula of the product performance index α is as follows:

α=exp[（tm-TM）/TM+（tz-TZ）/TZ]+sin[lg（fl/FL）];

Among them, tm is the machine cycle time of the designed product, TM is the rated value of the machine cycle of the designed product, tz is the instruction cycle time of the designed product, TZ is the rated value of the product instruction cycle, fl is the number of floating-point operations per second of the product, and FL is the rated value of the number of floating-point operations per second of the product.

Furthermore, the performance label analysis unit compares the product performance index α with each preset performance index, and analyzes the performance label of the product according to the comparison result and the product category analysis result, wherein:

When the product category is a fixed scene label, if α＜T1, the performance label analysis unit does not set a performance label for the designed product; if α≥T1, the performance label analysis unit sets a performance label for the designed product;

When the product category is a performance scenario label, if α＜T2, the performance label analysis unit does not set a performance label for the designed product; if α≥T2, the performance label analysis unit sets a performance label for the designed product;

Among them, T1 is the minimum scenario performance index, T2 is the average performance index of similar products, and T1≤T2.

Furthermore, the product feature analysis module is further provided with a quality label analysis unit, and the quality label analysis unit analyzes the product quality label according to the product category analysis result and the product specification information;

The quality label analysis unit calculates the product quality index β according to the product specification information. The calculation formula of the product quality index β is as follows:

β = ln{1 + [(t-T)/T]};

Among them, t is the shelf life of the product, and T is the average shelf life of similar products;

The quality label analysis unit compares the product quality index β with each preset quality index, and analyzes the product quality label according to the comparison result and the product category analysis result, wherein:

When the product category is a fixed scene label, if β＜B1, the quality label analysis unit does not set a quality label for the design product; if β≥B1, the quality label analysis unit sets a quality label for the design product;

When the product category is a performance scenario label, if β＜B2, the quality label analysis unit does not set a quality label for the design product; if β≥B2, the quality label analysis unit sets a quality label for the design product;

Among them, B1 is the minimum scene performance index, B2 is the average quality index of similar products, B1≤B2.

Furthermore, the product feature analysis module is further provided with a product feature analysis unit, which analyzes the product features according to the performance label analysis results, the quality label analysis results and the product category analysis results, wherein:

When the product category is a fixed-scenario product, if a1×(α-Ti)/Ti+a2×(β-Bj)/Bj＜K1, the product feature analysis unit sets the feature vector of the designed product to η1, and sets η1=(α, β); if a1×(α-Ti)/Ti+a2×(β-Bj)/Bj≥K1, the product feature analysis unit sets the feature vector of the designed product to η2, and sets η2=(α×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K1]/K1, β×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K1]/K1);

When the product category is a performance product, if a3×(α-Ti)/Ti+a4×(β-Bj)/Bj＜K2, the product feature analysis unit sets the feature vector of the designed product to η3, setting η3=(α, β); if a3×(α-Ti)/Ti+a4×(β-Bj)/Bj≥K2, the product feature analysis unit sets the feature vector of the designed product to η4, setting η4=(α×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K2]/K2, β×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K2]/K2);

Among them, a1 is the performance weight of fixed-scenario products, a2 is the quality weight of fixed-scenario products, a3 is the performance weight of performance products, a4 is the quality weight of performance products, K1 is the first preset feature index, K2 is the second preset feature index, K1＜K2, i=1,2 and j=1,2.

Furthermore, the user analysis module is provided with a user state analysis unit, and the user state analysis unit analyzes the user selection state according to the user information;

The user state analysis unit extracts functional keywords of historically selected products according to product functional information of historically selected products by the user, and divides the historically selected products into categories according to the functional keywords of the historically selected products;

The user state analysis unit sets the user selection state vector γ according to the product category classification result of the historically selected products, and sets γ=(N1, N2);

Among them, N1 is the number of products whose product category of historical selection products is fixed scenario products, and N2 is the number of products whose product category of historical selection products is performance products;

The user state analysis unit analyzes the user selection state according to the user selection state vector γ, wherein:

when ＜H and When the user state analysis unit determines that the user selection state is a fuzzy scene selection state;

when ≥H and When the user state analysis unit determines that the user selection state is a priority scene selection state;

when ＜H and When , the user state analysis unit determines that the user selection state is a fuzzy performance selection state;

when ≥H and When the user state analysis unit determines that the user selection state is a priority performance selection state;

Where n is a unit vector, n=(1,0) is set, H is the modulus of the preset user selection state, θ1 is the preset vector angle and 0°＜θ1≤45°.

Furthermore, the user analysis module is further provided with an adjustment unit, which calculates the average performance index Pα of historically selected products and sets Pα=[α(1)+α(2)+...+α(N)]/N;

Where α(1) is the product performance index of the first historically selected product, α(2) is the product performance index of the second historically selected product, α(N) is the product performance index of the Nth historically selected product, and N is the number of historically selected products;

The adjustment unit adjusts the analysis process of the user selection state according to the average performance index Pα of the historically selected products, wherein:

When Pα＜T2, the adjustment unit determines that the user's demand for product performance is normal and does not make any adjustment;

When Pα≥T2, the adjustment unit determines that the user has high requirements for product performance, and adjusts the user selection state vector to γ', setting γ'=(N1,N2×{1+sin[(T2-Pα)/Pα]}).

Furthermore, the user analysis module is further provided with an optimization unit, which compares the browsing time sc(z) of each historically selected product by the user with the preset time SC, and optimizes the analysis process of the user selection state according to the comparison result, setting z=1,2...N, wherein:

When sc(z)＜SC, the optimization unit determines that the browsing time of the historically selected product by the user is short, and optimizes the product performance index of the historically selected product to α'(z), setting α'(z)=α(z)×cos{[SC-sc(z)]/SC};

When sc(z)≥SC, the optimization unit determines that the browsing time of the user for the historically selected products is normal and does not perform optimization.

Furthermore, the correlation analysis module analyzes the correlation between the designed product and the user according to the user selection state analysis result and the product feature analysis result;

The correlation analysis module analyzes the correlation between the design product and the user according to the feature vector ηc of the design product and the user selection state, and sets c=1,2,3,4, where:

When the user selection state is a fuzzy scene selection state, if d1×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r1×|ηc|/|γ|≥Q1, the correlation analysis module determines that the design product is normally correlated with the user; if d1×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r1×|ηc|/|γ|＜Q1, the correlation analysis module determines that the design product is weakly correlated with the user;

When the user selection state is the priority scene selection state, if d2×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r2×|ηc|/|γ|≥Q1, the correlation analysis module determines that the design product is strongly correlated with the user; if d2×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r2×|ηc|/|γ|＜Q1, the correlation analysis module determines that the design product is normally correlated with the user;

When the user selection state is a fuzzy performance selection state, if d3×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r3×|ηc|/|γ|≥Q2, the correlation analysis module determines that the design product is normally correlated with the user; if d3×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r3×|ηc|/|γ|＜Q2, the correlation analysis module determines that the design product is weakly correlated with the user;

When the user selection state is the priority performance selection state, if d4×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r4×|ηc|/|γ|≥Q2, the correlation analysis module determines that the design product is strongly correlated with the user; if d4×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r4×|ηc|/|γ|＜Q2, the correlation analysis module determines that the design product is weakly correlated with the user;

Among them, Q1 is the first preset correlation coefficient, Q2 is the second preset correlation coefficient, Q1＜Q2, d1 is the first vector direction correlation coefficient weight, r1 is the first vector magnitude correlation coefficient weight, d2 is the second vector direction correlation coefficient weight, r2 is the second vector magnitude correlation coefficient weight, d3 is the third vector direction correlation coefficient weight, r3 is the third vector magnitude correlation coefficient weight, d4 is the fourth vector direction correlation coefficient weight, r4 is the fourth vector magnitude correlation coefficient weight, d2＞d1＞d3＞d4, r4＞r3＞r1＞r2, θ2 is the preset direction correlation angle 0°＜θ2＜15°.

Furthermore, the information classification module matches and classifies the design products according to the correlation analysis results between the design products and the users, wherein:

When the correlation between the design product and the user is strong, the information classification module lists the design product as a first-class product;

When the correlation between the design product and the user is normal, the information classification module classifies the design product as a second-category product;

When the correlation between the design product and the user is weak, the information classification module classifies the design product into three categories of products.

Compared with the prior art, the beneficial effects of the present invention lie in that the information required for the present embodiment is acquired by the information acquisition module, thereby improving the accuracy and completeness of information acquisition; the categories of the designed products are divided by the function classification module, thereby improving the accuracy of the functional division of the designed products; the product feature analysis module analyzes the product features, thereby improving the accuracy of the product feature analysis; the user analysis module analyzes the user selection status, thereby improving the accuracy of the user selection status analysis; the correlation analysis module analyzes the correlation between the designed product and the user, thereby improving the accuracy of the correlation analysis between the designed product and the user; the information classification module matches and classifies the designed products according to the correlation analysis results between the designed products and the users, thereby improving the accuracy of matching and classifying the designed products; the classification storage module stores the product identification information of the designed products as the matching classification results of the designed products, thereby improving the classification efficiency of the product design information.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of the structure of the product design information classification system of this embodiment.

FIG. 2 is a schematic diagram of the structure of the function classification module of this embodiment.

FIG3 is a schematic diagram of the structure of the product feature analysis module of this embodiment.

FIG. 4 is a schematic diagram of the structure of the user analysis module of this embodiment.

Detailed ways

In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

It should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" 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 a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

Please refer to FIG. 1, which is a schematic diagram of the structure of the product design information classification system of this embodiment, including:

The information acquisition module is used to acquire product function information, product specification information and product identification information of the designed product, and is also used to acquire user information; the product function information includes a product function description text of the designed product; the product specification information includes a machine cycle duration of the designed product, a machine cycle rating of the designed product, an instruction cycle duration of the designed product, a product instruction cycle rating, a number of floating-point operations per second of the product, a number of floating-point operations per second of the product rating and a product shelf life; the user information includes product function information and product specification information of the user's historically selected products; the present embodiment does not specifically limit the method for acquiring the product function information, product specification information and product identification information of the designed product, and those skilled in the art can freely set it, as long as the acquisition requirements of the product function information, product specification information and product identification information of the designed product are met, such as being acquired through a product design information database; the method for acquiring user information in the present embodiment is to acquire it through a programming language;

A function classification module, used to analyze product categories according to product function information of the designed product, the function classification module is connected to the information acquisition module;

A product feature analysis module, used to analyze product labels according to the product category analysis results and product specification information, and also used to analyze product features according to the product label analysis results, the product feature analysis module is connected to the function classification module;

A user analysis module, used to analyze the user selection status according to the user information, the user analysis module is connected to the information acquisition module;

A correlation analysis module, used to analyze the correlation between the designed product and the user according to the user selection state analysis result and the product feature analysis result, and the correlation analysis module is connected with the user analysis module and the product feature analysis module;

An information classification module, used for matching and classifying the design products according to the correlation analysis results between the design products and the users, the information classification module is connected to the correlation analysis module;

The classification storage module is used to classify and store the designed products according to the matching classification results of the designed products, and the classification storage module is connected to the information classification module.

Please refer to FIG. 2 , which is a schematic diagram of the structure of the function classification module of this embodiment, including:

A text processing unit, used to extract keywords from the product function description text of the designed product to obtain function keywords;

The product function analysis unit is used to analyze the product category according to the function keyword analysis result, and the product function analysis unit is connected to the text processing unit.

Please refer to FIG3 , which is a schematic diagram of the structure of the product feature analysis module of this embodiment, including:

A performance label analysis unit, used to analyze product performance labels based on product category analysis results and product specification information;

A quality label analysis unit, used to analyze product quality labels based on product category analysis results and product specification information;

The product feature analysis unit is used to analyze product features according to the performance label analysis results, the quality label analysis results and the product category analysis results. The product feature analysis unit is connected to the performance label analysis unit and the quality label analysis unit.

Please refer to FIG4, which is a schematic diagram of the structure of the user analysis module of this embodiment, including:

A user status analysis unit, used to analyze the user selection status according to the user information;

An adjustment unit, used to adjust the analysis process of the user selection state according to the average performance index of the historically selected products, the adjustment unit being connected to the user state analysis unit;

The optimization unit is used to optimize the analysis process of the user selection state according to the browsing time of the historically selected products, and the optimization unit is connected to the adjustment unit.

Specifically, the product design information classification system described in this embodiment is applied to the classification of electronic product design information; this embodiment does not specifically limit the categories of electronic products, and technical personnel in this field can set them freely. The electronic products in this embodiment can be smart electronic devices such as mobile phones, tablet computers, etc.; the purpose of classifying product design information described in this embodiment is to design a classification mode that better meets user needs; this embodiment analyzes product features and user selection status, and then analyzes the correlation between the designed product and the user based on the analysis results of the product features and user selection status, and classifies and stores the product design information based on the correlation analysis results, thereby improving the classification efficiency of product design information.

Specifically, the text processing unit performs text preprocessing on the product function information of the designed product;

The text processing unit extracts keywords from the product function description text of the designed product to obtain function keywords.

It is worth noting that the present embodiment does not specifically limit the process of extracting functional keywords based on the product function description text, and those skilled in the art can freely set it, and it only needs to meet the needs of extracting functional keywords, such as obtaining text phrases by performing text segmentation on the product function information, and assigning word vectors to the text phrases according to the text phrases and the pre-trained word vector file, and then performing cluster analysis on the assigned text phrases to obtain functional keywords; the present embodiment does not specifically limit the process of text segmentation, and those skilled in the art can freely set it, and it only needs to meet the needs of text segmentation, such as performing word segmentation on the product function description text through a stop word list and the mmseg4j tool; wherein, the stop word list and the pre-trained word vector file are obtained by obtaining them from the GitHub website.

Specifically, the text processing unit obtains function keywords by preprocessing the product function description text of the designed product, thereby improving the accuracy of function keyword analysis, thereby improving the accuracy of product classification, thereby improving the accuracy of product feature analysis, and ultimately improving the classification efficiency of product design information.

Specifically, the product function analysis unit analyzes the product category according to the function keyword analysis result of the designed product, wherein:

When the function keyword of the designed product belongs to the scene description keyword library, the product analysis unit determines that the product category of the product is a fixed scene product;

When the function keyword of the designed product belongs to the performance description keyword library, the product analysis unit determines that the product category of the product is a performance product.

It is worth noting that the present embodiment does not specifically limit the method for obtaining the "scene description keyword library" and the "performance description keyword library". Those skilled in the art can freely set them as long as the acquisition requirements of the "scene description keyword library" and the "performance description keyword library" are met, such as being obtained by user interactive input.

Specifically, the product function analysis unit analyzes the product category through the functional keyword analysis results of the designed product, thereby improving the accuracy of product classification, thereby improving the accuracy of product feature analysis, and ultimately improving the classification efficiency of product design information.

Specifically, the product feature analysis module is provided with a performance label analysis unit, and the performance label analysis unit analyzes the product performance label according to the product category analysis result and the product specification information;

The performance label analysis unit analyzes the performance label of the product according to the product category analysis result and the product specification information;

The performance label analysis unit calculates the product performance index α according to the product specification information. The calculation formula of the product performance index α is as follows:

α=exp[（tm-TM）/TM+（tz-TZ）/TZ]+sin[lg（fl/FL）];

Wherein, tm is the machine cycle time of the designed product, TM is the rated value of the machine cycle of the designed product, tz is the instruction cycle time of the designed product, TZ is the rated value of the instruction cycle of the product, fl is the number of floating-point operations per second of the product, and FL is the rated value of the number of floating-point operations per second of the product;

The performance label analysis unit compares the product performance index α with each preset performance index, and analyzes the product performance label according to the comparison result and the product category analysis result, wherein:

When the product category is a fixed scene label, if α＜T1, the performance label analysis unit does not set a performance label for the designed product; if α≥T1, the performance label analysis unit sets a performance label for the designed product;

When the product category is a performance scenario label, if α＜T2, the performance label analysis unit does not set a performance label for the designed product; if α≥T2, the performance label analysis unit sets a performance label for the designed product;

Among them, T1 is the minimum scenario performance index, T2 is the average performance index of similar products, and T1≤T2.

Specifically, the performance label analysis unit improves the accuracy of setting the product performance label by analyzing the product performance index of the designed product, thereby improving the accuracy of the product feature analysis, thereby improving the accuracy of the correlation analysis between the designed product and the user, and ultimately improving the classification efficiency of the product design information; it can be understood that this embodiment does not make specific limitations on the value of the minimum scenario performance index T1, and technical personnel in this field can set it freely, as long as the value requirements of the minimum scenario performance index T1 are met, such as the minimum scenario performance index T1 can be set to e, where e is specifically the natural logarithm; at the same time, this embodiment does not make specific limitations on the calculation process of the average performance index T2 of similar products, and its specific calculation process is to calculate the product performance index of similar products and take the average value thereof.

The product feature analysis module is further provided with a quality label analysis unit, which analyzes the product quality label according to the product category analysis result and the product specification information;

The quality label analysis unit calculates the product quality index β according to the product specification information. The calculation formula of the product quality index β is as follows:

β = ln{1 + [(t-T)/T]};

Among them, t is the shelf life of the product, and T is the average shelf life of similar products;

The quality label analysis unit compares the product quality index β with each preset quality index, and analyzes the product quality label according to the comparison result and the product category analysis result, wherein:

When the product category is a fixed scene label, if β＜B1, the quality label analysis unit does not set a quality label for the design product; if β≥B1, the quality label analysis unit sets a quality label for the design product;

When the product category is a performance scenario label, if β＜B2, the quality label analysis unit does not set a quality label for the design product; if β≥B2, the quality label analysis unit sets a quality label for the design product;

Among them, B1 is the minimum scene performance index, B2 is the average quality index of similar products, B1≤B2.

Specifically, the quality label analysis unit improves the accuracy of setting product performance labels by analyzing the product quality index of the designed product, thereby improving the accuracy of product feature analysis, thereby improving the accuracy of correlation analysis between the designed product and the user, and ultimately improving the classification efficiency of product design information; it can be understood that this embodiment does not make specific limitations on the value of the minimum scenario performance index B1, and technical personnel in this field can set it freely, as long as the value requirement of the minimum scenario performance index B1 is met, such as the minimum scenario performance index B1 can be set to 0.6; at the same time, the calculation process of the average quality index B2 of similar products is the same as the calculation process of the average performance index T2 of similar products.

Specifically, the product feature analysis module is further provided with a product feature analysis unit, which analyzes the product features according to the performance label analysis results, the quality label analysis results and the product category analysis results, wherein:

When the product category is a fixed-scenario product, if a1×(α-Ti)/Ti+a2×(β-Bj)/Bj＜K1, the product feature analysis unit sets the feature vector of the designed product to η1, and sets η1=(α, β); if a1×(α-Ti)/Ti+a2×(β-Bj)/Bj≥K1, the product feature analysis unit sets the feature vector of the designed product to η2, and sets η2=(α×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K1]/K1, β×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K1]/K1);

When the product category is a performance product, if a3×(α-Ti)/Ti+a4×(β-Bj)/Bj＜K2, the product feature analysis unit sets the feature vector of the designed product to η3, setting η3=(α, β); if a3×(α-Ti)/Ti+a4×(β-Bj)/Bj≥K2, the product feature analysis unit sets the feature vector of the designed product to η4, setting η4=(α×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K2]/K2, β×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K2]/K2);

Among them, a1 is the performance weight of the fixed scenario product, a2 is the quality weight of the fixed scenario product, a3 is the performance weight of the performance product, a4 is the quality weight of the performance product, a1＜a2 and a1+a2=1, a3＞a4 and a3+a4=1, K1 is the first preset feature index, K2 is the second preset feature index, K1＜K2, i=1,2 and j=1,2.

Specifically, the product feature analysis unit improves the accuracy of product feature analysis by analyzing product features according to performance label analysis results, quality label analysis results and product category analysis results, thereby improving the accuracy of correlation analysis between designed products and users, and ultimately improving the classification efficiency of product design information; it can be understood that this embodiment does not make specific limitations on the values of the first preset feature index K1 and the second preset feature index K2, and technical personnel in this field can set them freely, as long as the value requirements of the first preset feature index K1 and the second preset feature index K2 are met, such as the first preset feature index K1 can be set to 1, and the second preset feature index K2 can be set to 1.5.

Specifically, the user analysis module is provided with a user state analysis unit, and the user state analysis unit analyzes the user selection state according to the user information;

The user state analysis unit extracts functional keywords of historically selected products according to product functional information of historically selected products by the user, and divides the historically selected products into categories according to the functional keywords of the historically selected products;

The process of extracting functional keywords and analyzing product categories of historically selected products by the user state analysis unit is the same as the process of extracting functional keywords and analyzing product categories of the above-mentioned designed products;

The user state analysis unit sets the user selection state vector γ according to the product category classification result of the historically selected products, and sets γ=(N1, N2);

Among them, N1 is the number of products whose product category of historical selection products is fixed scenario products, and N2 is the number of products whose product category of historical selection products is performance products;

The user state analysis unit analyzes the user selection state according to the user selection state vector γ, wherein:

when ＜H and When the user state analysis unit determines that the user selection state is a fuzzy scene selection state;

when ≥H and When the user state analysis unit determines that the user selection state is a priority scene selection state;

when ＜H and When , the user state analysis unit determines that the user selection state is a fuzzy performance selection state;

when ≥H and When the user state analysis unit determines that the user selection state is a priority performance selection state;

Where n is a unit vector, n=(1,0) is set, H is the modulus of the preset user selection state, θ1 is the preset vector angle and 0°＜θ1≤45°.

It is worth noting that this embodiment is the modulus of the user-selected state vector γ, and n·γ represents the dot product between the unit vector n and the user-selected state vector γ.

Specifically, the user state analysis unit improves the accuracy of the user selection state analysis by analyzing the user selection state according to the user information, thereby improving the accuracy of the correlation analysis between the design product and the user, thereby improving the accuracy of the matching classification of the design product, and ultimately improving the classification efficiency of the product design information; it can be understood that this embodiment does not make a specific limitation on the value of the module length H of the preset user selection state, and technical personnel in this field can set it freely, as long as the value requirement of the module length H of the preset user selection state is met, such as the module length H of the preset user selection state can be set to 1.414.

Specifically, the adjustment unit calculates the average performance index Pα of the historically selected products and sets Pα=[α(1)+α(2)+...+α(N)]/N;

Where α(1) is the product performance index of the first historically selected product, α(2) is the product performance index of the second historically selected product, α(N) is the product performance index of the Nth historically selected product, and N is the number of historically selected products;

The adjustment unit adjusts the analysis process of the user selection state according to the average performance index Pα of the historically selected products, wherein:

When Pα＜T2, the adjustment unit determines that the user's demand for product performance is normal and does not make any adjustment;

When Pα≥T2, the adjustment unit determines that the user has high requirements for product performance, and adjusts the user selection state vector to γ', setting γ'=(N1,N2×{1+sin[(T2-Pα)/Pα]}).

Specifically, the adjustment unit improves the accuracy of the user selection status analysis by adjusting the analysis process of the user selection status according to the average performance index of the historically selected products, thereby improving the accuracy of the correlation analysis between the designed products and the users, thereby improving the accuracy of the matching classification of the designed products, and ultimately improving the classification efficiency of the product design information.

Specifically, the optimization unit compares the browsing time sc(z) of each historically selected product by the user with the preset time SC, and optimizes the analysis process of the user selection state according to the comparison result, setting z=1,2...N, where:

When sc(z)＜SC, the optimization unit determines that the browsing time of the historically selected product by the user is short, and optimizes the product performance index of the historically selected product to α'(z), setting α'(z)=α(z)×cos{[SC-sc(z)]/SC};

When sc(z)≥SC, the optimization unit determines that the browsing time of the user for the historically selected products is normal and does not perform optimization.

Specifically, the optimization unit improves the accuracy of the user selection status analysis by optimizing the analysis process of the user selection status according to the browsing time of the historically selected products, thereby improving the accuracy of the correlation analysis between the designed products and the users, thereby improving the accuracy of the matching classification of the designed products, and ultimately improving the classification efficiency of the product design information; it can be understood that this embodiment does not make a specific limitation on the value of the preset time length SC, and it is only necessary to meet the value requirement of the preset time length SC, such as the preset time length SC can be set to 15s.

Specifically, the correlation analysis module analyzes the correlation between the designed product and the user according to the user selection state analysis result and the product feature analysis result;

The correlation analysis module analyzes the correlation between the design product and the user according to the feature vector ηc of the design product and the user selection state, and sets c=1,2,3,4, where:

When the user selection state is a fuzzy scene selection state, if d1×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r1×|ηc|/|γ|≥Q1, the correlation analysis module determines that the design product is normally correlated with the user; if d1×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r1×|ηc|/|γ|＜Q1, the correlation analysis module determines that the design product is weakly correlated with the user;

When the user selection state is the priority scene selection state, if d2×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r2×|ηc|/|γ|≥Q1, the correlation analysis module determines that the design product is strongly correlated with the user; if d2×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r2×|ηc|/|γ|＜Q1, the correlation analysis module determines that the design product is normally correlated with the user;

When the user selection state is a fuzzy performance selection state, if d3×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r3×|ηc|/|γ|≥Q2, the correlation analysis module determines that the design product is normally correlated with the user; if d3×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r3×|ηc|/|γ|＜Q2, the correlation analysis module determines that the design product is weakly correlated with the user;

When the user selection state is the priority performance selection state, if d4×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r4×|ηc|/|γ|≥Q2, the correlation analysis module determines that the design product is strongly correlated with the user; if d4×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r4×|ηc|/|γ|＜Q2, the correlation analysis module determines that the design product is weakly correlated with the user;

Among them, Q1 is the first preset correlation coefficient, Q2 is the second preset correlation coefficient, Q1＜Q2, d1 is the first vector direction correlation coefficient weight, r1 is the first vector magnitude correlation coefficient weight, d2 is the second vector direction correlation coefficient weight, r2 is the second vector magnitude correlation coefficient weight, d3 is the third vector direction correlation coefficient weight, r3 is the third vector magnitude correlation coefficient weight, d4 is the fourth vector direction correlation coefficient weight, r4 is the fourth vector magnitude correlation coefficient weight, d2＞d1＞d3＞d4, r4＞r3＞r1＞r2, θ2 is the preset direction correlation angle 0°＜θ2＜15°.

Specifically, the correlation analysis module improves the accuracy of the correlation analysis between the designed product and the user by analyzing the correlation between the designed product and the user according to the user selection status analysis results and the product feature analysis results, thereby improving the accuracy of the matching classification of the designed products, and ultimately improving the classification efficiency of the product design information; it can be understood that this embodiment does not make specific limitations on the values of the first preset correlation coefficient Q1 and the second preset correlation coefficient Q2, and it is only necessary to meet the value requirements of the preset correlation coefficient Q1 and the second preset correlation coefficient Q2, such as the preset correlation coefficient Q1 can be set to 0.6, and the second preset correlation coefficient Q2 can be set to 0.8; at the same time, df+rf=1, set f=1,2,3,4.

Specifically, the information classification module matches and classifies the design products according to the correlation analysis results between the design products and the users, wherein:

When the correlation between the design product and the user is strong, the information classification module lists the design product as a first-class product;

When the correlation between the design product and the user is normal, the information classification module classifies the design product as a second-category product;

When the correlation between the design product and the user is weak, the information classification module classifies the design product into three categories of products.

Specifically, the information classification module matches and classifies the design products according to the correlation analysis results between the design products and the users, thereby improving the accuracy of matching and classifying the design products, and ultimately improving the classification efficiency of the product design information.

It is worth noting that the first category of products described in this embodiment specifically refers to the first category of products being recommended and pushed to users when users search for this category of products; the second category of products specifically refers to the second category of products being recommended and pushed to users as substitutes for the first category of products when there is no first category of products; the third category of products specifically refers to non-user selected products and are not recommended and pushed to users.

Specifically, the classification storage module stores the product identification information of the designed product with the matching classification result of the designed product.

Specifically, the classification storage module improves the classification efficiency of product design information.

So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims (10)

1. A product design information classification system, characterized by comprising: An information acquisition module is used to acquire product function information, product specification information and product identification information of the designed product, and is also used to acquire user information; A function classification module is used to analyze the function keywords of the design product according to the product function information of the design product, and to analyze the product category according to the function keyword analysis result of the design product; A product feature analysis module, used to analyze the quality label and performance label of the designed product according to the product category analysis result and product specification information, and also used to analyze the product features according to the quality label and performance label analysis result of the designed product; User analysis module, used to analyze user selection status based on user information; A correlation analysis module is used to analyze the correlation between the designed product and the user based on the user selection status analysis results and the product feature analysis results; An information classification module is used to match and classify design products according to the correlation analysis results between design products and users; The classification storage module is used to classify and store the product identification information of the designed product according to the matching classification results of the designed product. 2. A product design information classification system according to claim 1, characterized in that the product feature analysis module is provided with a performance label analysis unit, and the performance label analysis unit analyzes the product performance label according to the product category analysis result and the product specification information; The performance label analysis unit analyzes the performance label of the product according to the product category analysis result and the product specification information; The performance label analysis unit calculates the product performance index α according to the product specification information. The calculation formula of the product performance index α is as follows: α=exp[（tm-TM）/TM+（tz-TZ）/TZ]+sin[lg（fl/FL）]; Among them, tm is the machine cycle time of the designed product, TM is the rated value of the machine cycle of the designed product, tz is the instruction cycle time of the designed product, TZ is the rated value of the product instruction cycle, fl is the number of floating-point operations per second of the product, and FL is the rated value of the number of floating-point operations per second of the product. 3. A product design information classification system according to claim 2, characterized in that the performance label analysis unit compares the product performance index α with each preset performance index, and analyzes the product performance label according to the comparison result and the product category analysis result, wherein: When the product category is a fixed scene label, if α＜T1, the performance label analysis unit does not set a performance label for the designed product; if α≥T1, the performance label analysis unit sets a performance label for the designed product; When the product category is a performance scenario label, if α＜T2, the performance label analysis unit does not set a performance label for the designed product; if α≥T2, the performance label analysis unit sets a performance label for the designed product; Among them, T1 is the minimum scenario performance index, T2 is the average performance index of similar products, and T1≤T2. 4. A product design information classification system according to claim 1, characterized in that the product feature analysis module is further provided with a quality label analysis unit, and the quality label analysis unit analyzes the product quality label according to the product category analysis result and the product specification information; The quality label analysis unit calculates the product quality index β according to the product specification information. The calculation formula of the product quality index β is as follows: β = ln{1 + [(t-T)/T]}; Among them, t is the shelf life of the product, and T is the average shelf life of similar products; The quality label analysis unit compares the product quality index β with each preset quality index, and analyzes the product quality label according to the comparison result and the product category analysis result, wherein: When the product category is a fixed scene label, if β＜B1, the quality label analysis unit does not set a quality label for the design product; if β≥B1, the quality label analysis unit sets a quality label for the design product; When the product category is a performance scenario label, if β＜B2, the quality label analysis unit does not set a quality label for the design product; if β≥B2, the quality label analysis unit sets a quality label for the design product; Among them, B1 is the minimum scene performance index, B2 is the average quality index of similar products, B1≤B2. 5. A product design information classification system according to claim 4, characterized in that the product feature analysis module is further provided with a product feature analysis unit, and the product feature analysis unit analyzes the product features according to the performance label analysis results, the quality label analysis results and the product category analysis results, wherein: When the product category is a fixed-scenario product, if a1×(α-Ti)/Ti+a2×(β-Bj)/Bj＜K1, the product feature analysis unit sets the feature vector of the designed product to η1, and sets η1=(α, β); if a1×(α-Ti)/Ti+a2×(β-Bj)/Bj≥K1, the product feature analysis unit sets the feature vector of the designed product to η2, and sets η2=(α×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K1]/K1, β×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K1]/K1); When the product category is a performance product, if a3×(α-Ti)/Ti+a4×(β-Bj)/Bj＜K2, the product feature analysis unit sets the feature vector of the designed product to η3, setting η3=(α, β); if a3×(α-Ti)/Ti+a4×(β-Bj)/Bj≥K2, the product feature analysis unit sets the feature vector of the designed product to η4, setting η4=(α×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K2]/K2, β×[2×a1×(α-Ti)/Ti+2×a2×(β-Bj)/Bj-K2]/K2); Among them, a1 is the performance weight of fixed-scenario products, a2 is the quality weight of fixed-scenario products, a3 is the performance weight of performance products, a4 is the quality weight of performance products, K1 is the first preset feature index, K2 is the second preset feature index, K1＜K2, i=1,2 and j=1,2. 6. A product design information classification system according to claim 1, characterized in that the user analysis module is provided with a user state analysis unit, and the user state analysis unit analyzes the user selection state according to the user information; The user state analysis unit extracts functional keywords of historically selected products according to product functional information of historically selected products by the user, and divides the historically selected products into categories according to the functional keywords of the historically selected products; The user state analysis unit sets the user selection state vector γ according to the product category classification result of the historically selected products, and sets γ=(N1, N2); Among them, N1 is the number of products whose product category of historical selection products is fixed scenario products, and N2 is the number of products whose product category of historical selection products is performance products; The user state analysis unit analyzes the user selection state according to the user selection state vector γ, wherein: when ＜H and When the user state analysis unit determines that the user selection state is a fuzzy scene selection state; when ≥H and When the user state analysis unit determines that the user selection state is a priority scene selection state; when ＜H and When , the user state analysis unit determines that the user selection state is a fuzzy performance selection state; when ≥H and When the user state analysis unit determines that the user selection state is a priority performance selection state; Where n is a unit vector, n=(1,0) is set, H is the modulus of the preset user selection state, θ1 is the preset vector angle and 0°＜θ1≤45°. 7. A product design information classification system according to claim 6, characterized in that the user analysis module is further provided with an adjustment unit, the adjustment unit calculates the average performance index Pα of historically selected products, and sets Pα=[α(1)+α(2)+...+α(N)]/N; Where α(1) is the product performance index of the first historically selected product, α(2) is the product performance index of the second historically selected product, α(N) is the product performance index of the Nth historically selected product, and N is the number of historically selected products; The adjustment unit adjusts the analysis process of the user selection state according to the average performance index Pα of the historically selected products, wherein: When Pα＜T2, the adjustment unit determines that the user's demand for product performance is normal and does not make any adjustment; When Pα≥T2, the adjustment unit determines that the user has high requirements for product performance, and adjusts the user selection state vector to γ', setting γ'=(N1,N2×{1+sin[(T2-Pα)/Pα]}). 8. A product design information classification system according to claim 7, characterized in that the user analysis module is further provided with an optimization unit, the optimization unit compares the browsing time sc(z) of each historically selected product by the user with the preset time SC, and optimizes the analysis process of the user selection state according to the comparison result, setting z=1,2...N, wherein: When sc(z)＜SC, the optimization unit determines that the browsing time of the historically selected product by the user is short, and optimizes the product performance index of the historically selected product to α'(z), setting α'(z)=α(z)×cos{[SC-sc(z)]/SC}; When sc(z)≥SC, the optimization unit determines that the browsing time of the user for the historically selected products is normal and does not perform optimization. 9. A product design information classification system according to claim 1, characterized in that the correlation analysis module analyzes the correlation between the design product and the user according to the user selection state analysis result and the product feature analysis result; The correlation analysis module analyzes the correlation between the design product and the user according to the feature vector ηc of the design product and the user selection state, and sets c=1,2,3,4, where: When the user selection state is a fuzzy scene selection state, if d1×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r1×|ηc|/|γ|≥Q1, the correlation analysis module determines that the design product is normally correlated with the user; if d1×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r1×|ηc|/|γ|＜Q1, the correlation analysis module determines that the design product is weakly correlated with the user; When the user selection state is the priority scene selection state, if d2×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r2×|ηc|/|γ|≥Q1, the correlation analysis module determines that the design product is strongly correlated with the user; if d2×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r2×|ηc|/|γ|＜Q1, the correlation analysis module determines that the design product is normally correlated with the user; When the user selection state is a fuzzy performance selection state, if d3×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r3×|ηc|/|γ|≥Q2, the correlation analysis module determines that the design product is normally correlated with the user; if d3×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r3×|ηc|/|γ|＜Q2, the correlation analysis module determines that the design product is weakly correlated with the user; When the user selection state is the priority performance selection state, if d4×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r4×|ηc|/|γ|≥Q2, the correlation analysis module determines that the design product is strongly correlated with the user; if d4×arccos[(ηc·γ)/(|ηc|×|γ|)]/θ2+r4×|ηc|/|γ|＜Q2, the correlation analysis module determines that the design product is weakly correlated with the user; Among them, Q1 is the first preset correlation coefficient, Q2 is the second preset correlation coefficient, Q1＜Q2, d1 is the first vector direction correlation coefficient weight, r1 is the first vector magnitude correlation coefficient weight, d2 is the second vector direction correlation coefficient weight, r2 is the second vector magnitude correlation coefficient weight, d3 is the third vector direction correlation coefficient weight, r3 is the third vector magnitude correlation coefficient weight, d4 is the fourth vector direction correlation coefficient weight, r4 is the fourth vector magnitude correlation coefficient weight, d2＞d1＞d3＞d4, r4＞r3＞r1＞r2, θ2 is the preset direction correlation angle 0°＜θ2＜15°. 10. A product design information classification system according to claim 9, characterized in that the information classification module matches and classifies the design products according to the correlation analysis results between the design products and the users, wherein: When the correlation between the design product and the user is strong, the information classification module lists the design product as a first-class product; When the correlation between the design product and the user is normal, the information classification module classifies the design product as a second-category product; When the correlation between the design product and the user is weak, the information classification module classifies the design product into three categories of products.