A supply and demand relationship quantitative evaluation method, device, equipment and medium

Abstract

Embodiments of the present specification provide a supply and demand relationship quantitative evaluation method, device, equipment and medium, wherein the method comprises: obtaining a supply side index set and a demand side index set of a target category; wherein the supply side index set and the demand side index set both contain indexes of a plurality of content carriers, and at least two indexes in the indexes of the plurality of content carriers are heterogeneous with each other; performing data standardization processing on each index in the supply side index set and the demand side index set to obtain each standardized index value; determining the weight of each index in the index set based on a factor analysis method according to the each standardized index value in the supply side index set and the demand side index set; determining a supply index and a demand index of the target category according to the weight and the each standardized index value; and determining a supply and demand balance coefficient of the target category according to the supply index and the demand index.

Description

Technical Field

[0001] The embodiments in this specification relate to the field of data analysis technology, and in particular to a method, apparatus, equipment and medium for quantitative evaluation of supply and demand relationships. Background Technology

[0002] In the current era of deep integration between e-commerce and content communities, commodity transactions and consumption behaviors are widely distributed across various content platforms and scenarios. Comprehensive assessment of supply and demand relationships for product categories has become a key requirement supporting platform operational decisions. Taking common e-commerce categories such as cosmetics and apparel as examples, their supply and demand data is scattered across different content platforms such as live streaming, e-commerce platforms, and product cards, and covers multiple user usage scenarios such as search and news feeds, leading to multiple technical challenges in supply and demand assessment.

[0003] Taking the "beauty" category as an example, its supply may be reflected simultaneously in influencer live streams, user-shared product reviews, and direct product displays, while user demand may be expressed through different behavioral paths such as keyword searches, clicks, or final purchases. Because the evaluation indicators for different platforms differ fundamentally in type, scale, and business meaning, it is difficult for platforms to construct a unified and objective quantitative system to measure the relationship between the category's supply capacity and actual demand.

[0004] Therefore, how to construct a method that can objectively and accurately quantify the supply and demand relationship at the category level in the complex e-commerce environment with multiple content carriers has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] In view of this, embodiments of this specification provide a method for quantitatively assessing supply and demand relationships. One or more embodiments of this specification also relate to a device for quantitatively assessing supply and demand relationships, a computing device, a computer-readable storage medium, and a computer program product, to address the technical deficiencies existing in the prior art.

[0006] According to a first aspect of the embodiments of this specification, a method for quantitatively assessing supply and demand relationships is provided, comprising: Obtain the supply-side indicator set and the demand-side indicator set of the target product category; wherein, both the supply-side indicator set and the demand-side indicator set contain indicators of multiple content carriers, and at least two of the indicators of the multiple content carriers are heterogeneous to each other. The indicators in the supply-side indicator set and the demand-side indicator set are standardized to obtain the values ​​of each standardized indicator. Based on factor analysis, the weight of each indicator in its respective indicator set is determined according to the standardized indicator values ​​in the supply-side indicator set and the demand-side indicator set. Based on the weights and the values ​​of each standardized indicator, the supply index and demand index of the target product category are determined. The supply-demand balance coefficient of the target product category is determined based on the supply index and the demand index.

[0007] According to a second aspect of the embodiments of this specification, a supply and demand relationship quantitative assessment device is provided, comprising: The acquisition module is configured to acquire a supply-side indicator set and a demand-side indicator set for the target product category; wherein, both the supply-side indicator set and the demand-side indicator set contain indicators of multiple content carriers, and at least two of the indicators of the multiple content carriers are heterogeneous to each other. The data processing module is configured to perform data standardization processing on each indicator in the supply-side indicator set and the demand-side indicator set to obtain the values ​​of each standardized indicator. The first determining module is configured to determine the weight of each indicator in its respective indicator set based on factor analysis, according to the standardized indicator values ​​of the supply-side indicator set and the demand-side indicator set. The second determining module is configured to determine the supply index and demand index of the target product category based on the weights and the values ​​of each standardized indicator. The third determining module is configured to determine the supply-demand balance coefficient of the target product category based on the supply index and the demand index.

[0008] According to a third aspect of the embodiments of this specification, a computing device is provided, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the above-described quantitative assessment method for supply and demand relationship.

[0009] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided that stores computer-executable instructions, which, when executed by a processor, implement the steps of the above-described quantitative assessment method for supply and demand relationships.

[0010] According to a fifth aspect of the embodiments of this specification, a computer program product is provided, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described quantitative assessment method for supply and demand relationship.

[0011] The quantitative assessment method for supply and demand relationships provided in the embodiments of this specification, after obtaining the supply-side indicator set and demand-side indicator set of the target product category, can perform data standardization processing on each indicator in the supply-side indicator set and demand-side indicator set to obtain the standardized indicator values. Both the supply-side indicator set and the demand-side indicator set contain indicators with multiple content carriers, and at least two of the indicators with multiple content carriers are heterogeneous. Then, based on factor analysis, the weight of each indicator in its respective indicator set is determined according to the standardized indicator values ​​in the supply-side indicator set and the demand-side indicator set. Based on the weight of each indicator in its respective indicator set and the standardized indicator values, the supply index and demand index of the target product category are determined. Finally, the supply-demand balance coefficient of the target product category can be determined based on the supply index and demand index. Therefore, in this embodiment, by incorporating heterogeneous indicators from multiple content carriers to construct supply-side and demand-side indicator sets, cross-carrier supply and demand data analysis is achieved. Combined with data standardization, the differences in dimensions and type conflicts of heterogeneous indicators are eliminated, providing a unified evaluation basis for multi-content carrier indicator data that were previously incomparable. This quantifies the platform's supply capacity and user demand across content carriers on a unified scale, automatically calculating the objective weights of each indicator to construct standardized supply and demand indices. This effectively assesses the platform's supply and demand health in the target category, significantly improving the comprehensiveness and consistency of supply and demand relationship assessments. Furthermore, this embodiment uses factor analysis instead of manual experience to determine indicator weights, avoiding biases caused by subjectively setting indicator weights. The indicator weights are determined by the standardized indicator values ​​in the supply-side and demand-side indicator sets, enabling dynamic responses to market supply and demand changes. Compared to the traditional fixed-weight model, this significantly improves the reliability, objectivity, and adaptability to real-world scenarios of supply and demand relationship assessment results. Attached Figure Description

[0012] Figure 1 This is an application scenario diagram of a quantitative assessment of supply and demand relationship provided by one embodiment of this specification; Figure 2 This is a flowchart illustrating a quantitative assessment method for supply and demand relationships provided in one embodiment of this specification; Figure 3 This is a flowchart illustrating the process of obtaining the indicator set in a quantitative assessment method for supply and demand relationships, provided in one embodiment of this specification. Figure 4 This is a flowchart of data standardization processing in a quantitative assessment method for supply and demand relationships provided in one embodiment of this specification; Figure 5 This is a flowchart of factor analysis in a quantitative assessment method for supply and demand relationships provided in one embodiment of this specification. Figure 6This is an overall flowchart of a quantitative assessment method for supply and demand relationships provided in one embodiment of this specification; Figure 7 This is a schematic diagram of the structure of a supply and demand relationship quantitative assessment device provided in one embodiment of this specification; Figure 8 This is a structural block diagram of a computing device provided in one embodiment of this specification. Detailed Implementation

[0013] Many specific details are set forth in the following description to provide a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0014] The terminology used in one or more embodiments of this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of this specification. The singular forms “a,” “described,” and “the” as used in one or more embodiments of this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0015] It should be understood that although the terms first, second, etc., may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first may also be referred to as second without departing from the scope of one or more embodiments of this specification, and similarly, second may also be referred to as first. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0016] Furthermore, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties. Moreover, the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0017] First, the terms and concepts used in one or more embodiments of this specification will be explained.

[0018] A Standardized Product Unit (SPU) is a core basic unit used in product management, e-commerce operations, and supply chain collaboration to standardize the classification and uniform identification of products. Essentially, a SPU defines a category of products with the same core attributes (such as brand, model, style, core function, etc.), distinguishing it from a Sales Unit (SKU) which represents differentiated attributes such as specific specifications, configurations, and colors. By extracting the common characteristics of products, SPUs achieve standardized management of product information, thereby improving the efficiency of product classification, retrieval, and inventory management. It is an important tool for achieving large-scale product operations.

[0019] Price Band Coverage (PBC) is a core analytical indicator used in retail, e-commerce, and other fields to evaluate the rationality of product pricing strategies. The core of PBC is to measure the completeness of product distribution within a pre-defined price range (such as low-end, mid-range, and high-end). By statistically analyzing the number, percentage, and sales contribution of products in each price segment, it determines whether product pricing covers the different spending power of the target customer group, thereby providing data support for optimizing product mix, adjusting pricing strategies, and increasing market penetration.

[0020] Product category (PC) is a systematic classification of products in retail, e-commerce, and supply chain management based on common characteristics such as core attributes, functions, consumption scenarios, and target customer groups. Product category is the basic unit of product management and operation. It helps companies to sort out product structure, optimize display layout, and develop differentiated marketing plans. It also enables consumers to more efficiently identify and retrieve the products they need. Different categories usually have clear boundaries and distinguishing standards, and can be further subdivided into more specific subcategories according to business needs.

[0021] Page views (PV) are a core metric in internet operations and traffic statistics. They refer to the number of times a user loads or refreshes a single page while visiting a website or mobile application. The statistical logic for page views is that multiple views or refreshes of the same page by the same user are cumulatively counted. Page views directly reflect the frequency of visits to a specific page and the reach of its content. They are often used to evaluate the attractiveness of website content, the effectiveness of marketing promotions, and user behavior paths, providing data support for optimizing operational strategies.

[0022] Page Views Click-Through Rate (PVCTR) is a composite traffic evaluation metric in internet operations and advertising. At its core, PVCTR uses page views (PV) as the exposure base and measures the page's appeal and conversion rate by calculating the ratio of clicks to total page views within a specific statistical period. This metric accurately reflects the click intent of the user group actually reached by the page and is often used to optimize page content layout, adjust ad placement, and provide data support for improving traffic conversion efficiency and operational decisions.

[0023] Unique Visitors (UV) is a core metric in internet traffic statistics and operational analysis. It refers to the number of individual users who visit a website, mobile application, mini-program, or other digital platform within a specific statistical period (such as a day, a week, or a month). The statistical logic for unique visitors involves deduplicating multiple visits from the same user. User uniqueness is typically identified through methods such as cookies, IP addresses, and user account identifiers. It directly reflects the platform's actual user reach and user base, serving as a key basis for assessing a platform's user growth potential, user stickiness, and market penetration effectiveness.

[0024] Unique Visitors Click-Through Rate (UVCTR) is a composite traffic conversion evaluation metric in internet operations and advertising. The core calculation logic of UVCTR is the ratio of the total number of clicks on target content, ads, or pages within a specific statistical period to the number of unique visitors (UV). Unlike PVCTR, which is based on page views (PV), this metric deduplicates user behavior, more accurately reflecting the click intent and interest level of individual users in target content. It is often used to evaluate the effectiveness of user segmentation operations, optimize targeted advertising strategies, and determine the appeal of content to core user groups, providing data support for improving traffic conversion quality and user value mining.

[0025] Factor analysis (FA) is a multivariate statistical analysis method based on the idea of ​​dimensionality reduction. Its core logic is to extract a few latent common factors from a set of observed variables with complex correlations that can reflect the internal relationships and common characteristics of the variables. These less dimensional common factors are used to explain most of the information covered by the original variables, thereby achieving the goal of simplifying the data structure and revealing the potential laws between variables. Factor analysis is widely used in data analysis and structure exploration in many fields such as psychology, economics, and sociology.

[0026] The Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) is a commonly used preliminary statistical test in factor analysis. The core of the KMO test is to quantitatively assess whether sample data is suitable for factor analysis by calculating the ratio of the simple correlation coefficient to the partial correlation coefficient between variables. The KMO value ranges from 0 to 1; the closer the value is to 1, the more common factors and the stronger the correlation between variables, making it more suitable for factor analysis. If the value is below 0.5, it indicates a weak correlation between variables, making it unsuitable for factor analysis. The KMO test is often used in conjunction with the Bartlett's test of sphericity to provide a scientific quantitative basis for determining the applicability of factor analysis.

[0027] Bartlett's Test of Sphericity (BTS) is a crucial preliminary statistical test in factor analysis. Its core purpose is to verify whether a significant correlation exists between the original variables being analyzed. Specifically, it tests whether the correlation coefficient matrix of the variables significantly deviates from the identity matrix. If the p-value is less than the preset significance level (e.g., 0.05), the null hypothesis of "no correlation between variables" is rejected, indicating that the variables are suitable for factor analysis. Conversely, a p-value greater than the p-value indicates strong independence of the variables, making them unsuitable for extracting common factors. The Bartlett's Test of Sphericity provides a critical prerequisite for ensuring the validity and reliability of factor analysis results.

[0028] Principal Component Analysis (PCA) is a classic multivariate statistical dimensionality reduction method. Its core logic is to transform a set of correlated high-dimensional variables into a set of linearly independent low-dimensional variables (principal components) through orthogonal transformation. These principal components are ordered from highest to lowest variance contribution rate. The first principal component retains the most information from the original data, and subsequent principal components decrease in size and are independent of each other. This simplifies the data dimensionality while preserving the core features of the original data to the greatest extent possible. PCA is widely used in data preprocessing and feature extraction in fields such as data analysis, pattern recognition, machine learning, and image processing.

[0029] Common Factor (CF) is a core latent variable in factor analysis. It refers to an abstract dimension that can explain the common variation characteristics among a set of correlated observed variables. Common factors are not directly observed, but are derived through correlation analysis of multiple observed variables. They can reflect the inherent common laws of observed variables and are widely used in multivariate statistical analysis, machine learning feature extraction and other scenarios. They are a key element in achieving data dimensionality reduction and revealing the latent structure of data.

[0030] The Varimax Method (Varimax) is a commonly used orthogonal factor rotation method in factor analysis. The core logic of the Varimax Method is to maximize the variance of each common factor loading value by orthogonally transforming the initial factor loading matrix, so that the loading values ​​are polarized towards 0 and 1, and each common factor is strongly correlated with only a few observed variables. This simplifies the meaning of the common factors, improves the interpretability of each common factor, and helps analysts to more clearly identify the actual meaning of each common factor.

[0031] Variance Contribution Rate (VCR) is a core evaluation metric in multivariate statistical dimensionality reduction methods such as principal component analysis and factor analysis. It refers to the proportion of the total variance of the original variables that a single principal component or common factor can explain. It is calculated by dividing the variance of that component or factor by the total variance of the original variables. A higher VCR value directly reflects the degree to which the corresponding principal component or common factor retains information from the original data; a higher value indicates that the common factor covers richer features of the original data. VCR is often used to screen key principal components or common factors, providing a quantitative basis for subsequent data dimensionality reduction and analytical decisions.

[0032] The Supply Index (SI) is a core quantitative indicator in economics, industry analysis, and supply chain management. By integrating key supply-side factors such as production capacity, inventory levels, output data, and delivery cycles within a market or industry, the SI constructs a standardized calculation model to measure the supply capacity and level of products or services within a specific period. It can intuitively reflect the sufficiency, fluctuation trends, and potential gaps on the supply side, providing crucial data support for assessing market supply and demand balance, formulating production scheduling strategies, and adjusting industrial supply policies.

[0033] The Demand Index (DI) is a core analytical indicator used in economics, market research, e-commerce operations, and other fields to quantitatively assess the degree and trend of market demand for a particular product or service within a specific period. The Demand Index typically integrates multi-dimensional demand-side information such as user search volume, order inquiries, potential purchase intentions, and historical consumption data. It is calculated using a standardized model to produce a relative value. A high or low DI value directly reflects market demand and provides scientific data support for businesses to analyze market trends, adjust production plans, optimize inventory layout, and formulate marketing strategies.

[0034] In the current platform environment where e-commerce and content communities are integrated, transactions and consumption occur in complex scenarios where multiple content carriers (such as live streaming, product notes, and product cards) and multiple venues (such as news feeds and search) coexist. Existing technologies have the following main problems: 1. Heterogeneous indicators across different content carriers make horizontal comparison impossible: The evaluation indicators used for different content carriers (such as live streaming, product notes, product cards, etc.) vary greatly, making it difficult to unify and quantify them using traditional methods.

[0035] 2. Supply-side indicators are disconnected from demand-side indicators: The supply side usually focuses on content distribution volume, product inventory, etc., while the demand side focuses on search volume, click volume, etc. There is a lack of coupling mechanism to judge the matching degree between "effective supply" and "effective demand".

[0036] 3. Weight allocation relies on human experience: When constructing a comprehensive evaluation model, the weights of various indicators are usually set subjectively by experts, lacking data support and making it difficult to dynamically adapt to market changes, leading to biased evaluation results.

[0037] To address the aforementioned problems, this specification provides a method for quantitatively assessing supply and demand relationships. One or more embodiments of this specification also relate to a device for quantitatively assessing supply and demand relationships, a computing device, a computer-readable storage medium, and a computer program product, to solve the problems existing in the prior art.

[0038] The quantitative assessment method for supply and demand relationships provided in one embodiment of this specification can be applied to the server side of a supply and demand relationship assessment platform. (See also...) Figure 1 , Figure 1 This diagram illustrates an application scenario for quantitative assessment of supply and demand relationships, provided by one embodiment of this specification.

[0039] like Figure 1As shown, user terminal 101 can watch product live streams, browse product notes, browse product cards, and conduct product transactions on e-commerce platform server terminal 102. E-commerce platform server terminal 102 can statistically record user terminal 101's browsing behavior, click behavior, number of views, and viewing duration. E-commerce platform server terminal 102 can also statistically analyze indicators such as the number of stores and live streams under a specific target product category, ultimately obtaining the supply-side indicator set and demand-side indicator set for the target product category. Both the supply-side and demand-side indicator sets can contain indicators from multiple content carriers, and at least two of the indicators from these multiple content carriers are heterogeneous. Supply-demand relationship assessment platform server terminal 103 can obtain the supply-side and demand-side indicator sets for the target product category from e-commerce platform server terminal 102, and perform data standardization processing on each indicator in the supply-side and demand-side indicator sets to obtain standardized indicator values. Subsequently, the supply and demand relationship assessment platform server 103 can determine the weight of each indicator in its respective indicator set based on factor analysis, according to the standardized indicator values ​​of each indicator in the supply-side indicator set and the demand-side indicator set. Based on the weight of each indicator in its respective indicator set and the standardized indicator values, the supply index and demand index of the target product category can be determined. Finally, the supply and demand balance coefficient of the target product category can be determined based on the supply index and the demand index.

[0040] In practical applications, after the supply and demand relationship assessment platform server 103 determines the supply and demand balance coefficient of the target product category, it can also generate instruction information for adjusting the supply and demand relationship of the target product category based on the supply and demand balance coefficient, so as to assist relevant personnel in making decisions on adjusting the supply and demand relationship of the target product category.

[0041] Optionally, the user terminal 101 can be any type of stationary or mobile computing device, including mobile computers or mobile computing devices (e.g., tablet computers, personal digital assistants, laptop computers, object content computers, netbooks, etc.), mobile phones (e.g., smartphones), wearable computing devices (e.g., smartwatches, smart glasses, etc.) or other types of mobile devices, or stationary computing devices such as desktop computers or personal computers.

[0042] Optionally, the e-commerce platform server 102 can be a standalone server, or a server network or server cluster, including but not limited to computers, network hosts, single network servers, multiple network server sets, or cloud servers composed of multiple servers. The cloud server can consist of a large number of computers or network servers based on cloud computing.

[0043] Optionally, the supply and demand relationship assessment platform server 103 can be a standalone server, or a server network or server cluster composed of servers, including but not limited to computers, network hosts, single network servers, multiple network server sets, or cloud servers composed of multiple servers. Among them, cloud servers can be composed of a large number of computers or network servers based on cloud computing.

[0044] In practical applications, data transmission can be conducted between the user terminal 101 and the e-commerce platform server 102, and between the e-commerce platform server 102 and the supply and demand relationship assessment platform server 103, through local area network connections, wide area network connections, Internet connections, or other types of data network connections, or through other means, without specific limitations.

[0045] Figure 1 The method described in the text involves the supply-demand relationship assessment platform obtaining the supply-side and demand-side indicator sets for the target product category. The platform then standardizes the data for each indicator in both sets, obtaining standardized indicator values. Both the supply-side and demand-side indicator sets contain indicators across multiple content carriers, with at least two of these carriers being heterogeneous. Factor analysis is then used to determine the weight of each indicator within its respective indicator set based on the standardized values. Based on these weights and standardized values, the supply and demand indices for the target product category are determined. Finally, the supply-demand balance coefficient for the target product category is calculated using these indices. Therefore, in this embodiment, by incorporating heterogeneous indicators from multiple content carriers to construct supply-side and demand-side indicator sets, cross-carrier supply and demand data analysis is achieved. Combined with data standardization, the differences in dimensions and type conflicts of heterogeneous indicators are eliminated, providing a unified evaluation basis for multi-content carrier indicator data that were previously incomparable. This quantifies the platform's supply capacity and user demand across content carriers on a unified scale, automatically calculating the objective weights of each indicator to construct standardized supply and demand indices. This effectively assesses the platform's supply and demand health in the target category, significantly improving the comprehensiveness and consistency of supply and demand relationship assessments. Furthermore, this embodiment uses factor analysis instead of manual experience to determine indicator weights, avoiding biases caused by subjectively setting indicator weights. The indicator weights are determined by the standardized indicator values ​​in the supply-side and demand-side indicator sets, enabling dynamic responses to market supply and demand changes. Compared to the traditional fixed-weight model, this significantly improves the reliability, objectivity, and adaptability to real-world scenarios of supply and demand relationship assessment results.

[0046] See Figure 2 , Figure 2 A flowchart of a quantitative assessment method for supply and demand relationship provided in one embodiment of this specification is shown, which specifically includes the following steps.

[0047] Step S202: Obtain the supply-side indicator set and demand-side indicator set of the target product category; wherein, both the supply-side indicator set and the demand-side indicator set contain indicators of multiple content carriers, and at least two of the indicators of the multiple content carriers are heterogeneous to each other.

[0048] The quantitative assessment method for supply and demand relationships provided in the embodiments of this specification can be applied to scenarios where the supply and demand relationships of goods on e-commerce platforms or other platforms that provide goods for sale are analyzed. During the analysis, it can eliminate the differences in the dimensions and type conflicts of heterogeneous indicators, so that the indicator data of multiple content carriers that were originally not directly comparable have a unified assessment basis. It realizes the quantification of the platform's supply capacity and user demand across content carriers on a unified dimension. It can automatically calculate the objective weight of each indicator, and then construct a standardized supply index and demand index, effectively assessing the health of the platform's supply and demand in the target category, and effectively improving the comprehensiveness and accuracy of the supply and demand relationship assessment.

[0049] In this context, the target category can refer to a set of products within the platform's product classification system that share common characteristics and require separate supply and demand analysis. Examples include: women's clothing, men's clothing, cosmetics, fresh produce, etc.

[0050] The supply-side indicator set may include indicators related to the supply side, which can be used to quantitatively evaluate the platform's overall supply capacity and level of goods, content, and services provided in the target category. The demand-side indicator set may include indicators related to the demand side, which can be used to quantitatively evaluate the overall intensity and scale of users' interest, intent, and purchasing behavior towards goods in the target category. Since subsequent embodiments in this specification will describe in detail the specific indicators included in the supply-side and demand-side indicator sets, they will not be repeated here.

[0051] The content carrier refers to the specific media format in which product information is presented on the platform. Content carriers mainly include, but are not limited to: live streaming (real-time video explanations), product notes (text / image / short video sharing), and product cards (static product information pages). Content carrier metrics can refer to indicators related to a specific content carrier. For example, the metric "Live Stream Exposure Page Views" is related to the content carrier "Live Stream"; the metric "Product Note Unique Visitor Click-Through Rate" is related to the content carrier "Product Notes"; and the metric "Product Card Click Count" is related to the content carrier "Product Card".

[0052] In the embodiments of this specification, heterogeneity can refer to fundamental differences between various indicators in terms of data type, statistical units, business meaning, and numerical distribution characteristics. For example, the units of measurement of various indicators may differ, such as from decimals of 0-1 (e.g., conversion rate) to integers in the millions or tens of millions (e.g., impressions). Another example is that the directionality of indicators may differ: a higher value for a positive indicator (e.g., page views) is better, while a higher value for a negative indicator (e.g., return rate) is worse.

[0053] In practical applications, supply and demand relationship assessment platforms can use various technical means such as data interface calls, database queries, log parsing, or receiving transmissions from external systems to obtain various indicators from the e-commerce platform's business database, data warehouse, or real-time data stream, including the supply-side indicator set and the demand-side indicator set.

[0054] It's important to note that in the complex platform ecosystem where content and e-commerce are deeply integrated, users' consumption decision-making process is highly fragmented. Information about products within the same category may reach users through various content formats such as live streams, text and image notes, and product cards. Users' interests and needs are also expressed through various behaviors such as searching, browsing, interacting, and purchasing. Relying solely on a single content format or a single-dimensional indicator for evaluation may yield biased results and fail to reflect the true supply and demand relationship of that product category from the platform's overall perspective.

[0055] In the embodiments of this specification, the constructed supply-side indicator set and demand-side indicator set both contain indicators of multiple content carriers, and these indicators include heterogeneous indicators. This overcomes the problem of limited evaluation perspective caused by the reliance on a single content carrier data source in traditional methods, so that the subsequently determined supply index and demand index can more realistically and completely reflect the supply and demand relationship of the target category in a complex platform ecosystem containing multiple content carriers.

[0056] Step S204: Perform data standardization processing on each indicator in the supply-side indicator set and the demand-side indicator set to obtain the standardized indicator values.

[0057] Data standardization refers to a series of mathematical transformations that convert raw, heterogeneous indicator data into standardized numerical sequences with uniform dimensions, consistent direction, and approximately following a standard normal distribution. The purpose of data standardization is to eliminate data incomparability caused by differences in indicator sources, types, and dimensions.

[0058] In practical applications, data standardization can include missing value imputation, data transformation, and standard normalization. Missing values ​​refer to numerical gaps in a specific sample for a particular indicator that are not recorded or cannot be calculated during data collection or integration. Data transformation ensures the consistency of indicator data direction, while standard normalization eliminates dimensional differences between indicators, ensuring that the processed indicator data are all of the same order of magnitude.

[0059] In the embodiments of this specification, there are heterogeneous indicators among the various indicators in the supply-side indicator set and the demand-side indicator set, which cannot be directly analyzed and processed using factor analysis. However, through data standardization, the differences in the dimensions and conflicts in the types of heterogeneous indicators can be eliminated, ensuring the consistency of the indicator directions. This enables the multi-content carrier indicator data that were originally not directly comparable to have a unified evaluation basis, realizing the quantification of platform supply capabilities and user needs across content carriers on a unified dimension. The standardized indicator values ​​obtained after data standardization can be analyzed and processed using factor analysis.

[0060] Step S206: Based on factor analysis, determine the weight of each indicator in its respective indicator set according to the standardized indicator values ​​in the supply-side indicator set and the demand-side indicator set.

[0061] In practical applications, before determining the weights of each indicator in its respective indicator set based on the standardized indicator values ​​in the supply-side and demand-side indicator sets using factor analysis, it's essential to first verify whether the standardized indicator data is suitable for factor analysis. Two common tests are: 1. KMO test: Calculate the Kaiser-Meyer-Olkin metric to compare the simple correlation coefficients and partial correlation coefficients between indicators. Generally, a KMO value greater than 0.6 indicates that the data is suitable for factor analysis. 2. Bartlett's test of sphericity: Test whether the correlation coefficient matrix is ​​an identity matrix (i.e., whether the indicators are independent). If the significance probability (p-value) is less than 0.05, the null hypothesis of indicator independence is rejected, and factor analysis is considered suitable. After passing these tests, factor analysis can be used to determine the weights of each indicator in its respective indicator set.

[0062] Specifically, based on factor analysis, the steps for determining the weight of each indicator in its respective indicator set based on the standardized indicator values ​​of the supply-side and demand-side indicator sets can include: 1. Extracting common factors: Using a validated standardized indicator matrix (such as a supply-side indicator matrix) as input, principal component analysis is used to extract common factors. The system calculates the eigenvalues ​​and eigenvectors of the correlation coefficient matrix and sorts them in descending order of eigenvalues, extracting components with eigenvalues ​​greater than 1 as common factors. 2. Factor rotation: The factor loading matrix is ​​rotated using the maximum variance method. After rotation, a new factor loading matrix is ​​obtained, maximizing the loading of each specific indicator on a certain common factor. Then, based on the set of indicators with high loadings on each common factor, business meanings (such as "content scale factor" or "monetization efficiency factor") can be assigned to the common factors. 3. Calculating indicator weights: Based on the rotated factor loading matrix and the variance contribution rate of the common factors, the weight of each indicator in its respective indicator set is calculated. The specific formula for calculating indicator weights can be:

[0063] in, This indicates the weight of index j; Indicates the number of common factors; This represents the variance contribution rate of the common factor k; This represents the loading of index j on common factor k.

[0064] In practical applications, the supply and demand relationship assessment platform can automatically perform the above-mentioned process of determining the weight of each indicator in its respective indicator set based on factor analysis on the supply-side indicator set and the demand-side indicator set respectively, and finally output two independent weight sets.

[0065] In the embodiments of this specification, the KMO test and Bartlett's test of sphericity can be used to analyze whether the data is suitable for factor analysis. If the test fails, factor analysis is not applicable. Thus, by conducting the test in advance, unsuitable data can be avoided from occupying computing resources, which is conducive to improving data processing efficiency and saving computing resources.

[0066] Furthermore, the common factors extracted by factor analysis and their variance contribution rates essentially reflect the main driving forces of supply and demand relationships within the current data period (e.g., the last 30 days). When the market environment or product category development stage changes, the variance contribution rates of common factors and the loadings of indicators will change accordingly, thus automatically adjusting the weight allocation. Based on factor analysis, the weights of each indicator in its respective indicator set are determined according to the standardized indicator values ​​in the supply-side and demand-side indicator sets. This ensures that the generation of each indicator weight is entirely based on the statistical characteristics of the indicator data itself, eliminating biases caused by differences in expert experience or personal preferences. It can dynamically respond to changes in market supply and demand, significantly improving the reliability, objectivity, and adaptability to real-world scenarios of supply and demand relationship assessment results compared to the traditional fixed-weight model.

[0067] Step S208: Determine the supply index and demand index of the target product category based on the weights and the values ​​of each standardized indicator.

[0068] Step S210: Determine the supply-demand balance coefficient of the target product category based on the supply index and the demand index.

[0069] In practical applications, after determining the weights of each supply-side indicator within its respective set of supply-side indicators using factor analysis, the supply index can be calculated based on these weights and standardized values. Similarly, after determining the weights of each demand-side indicator within its respective set of demand-side indicators using factor analysis, the demand index can be calculated based on these weights and standardized values.

[0070] In practical applications, after calculating the supply index and demand index of the target product category, the supply-demand balance coefficient can be calculated based on these indices. The core calculation logic of the supply-demand balance coefficient is the ratio of the demand index to the supply index. The supply-demand balance coefficient can intuitively reflect the supply-demand relationship: when the supply-demand balance coefficient is greater than a preset threshold (e.g., the preset threshold is 1), it indicates that market demand exceeds supply, and there is a supply gap; when the supply-demand balance coefficient is equal to the preset threshold, it represents that supply and demand are in an ideal equilibrium state; when the supply-demand balance coefficient is less than the preset threshold, it means that supply is excessive and demand is relatively insufficient.

[0071] In practical applications, since the supply and demand balance coefficient is a ratio calculated based on a standardized index, it eliminates the influence of the absolute size of the category. This allows the platform to fairly compare "home appliances" and "snacks" to determine which category is more in short supply, or to judge the changes in supply and demand for the same category before and after different promotional periods. This provides the platform with an objective and unified quantitative basis for overall resource coordination and priority ranking.

[0072] The quantitative assessment method for supply and demand relationships provided in the embodiments of this specification, after obtaining the supply-side indicator set and demand-side indicator set of the target product category, can perform data standardization processing on each indicator in the supply-side indicator set and demand-side indicator set to obtain the standardized indicator values. Both the supply-side indicator set and the demand-side indicator set contain indicators with multiple content carriers, and at least two of the indicators with multiple content carriers are heterogeneous. Then, based on factor analysis, the weight of each indicator in its respective indicator set is determined according to the standardized indicator values ​​in the supply-side indicator set and the demand-side indicator set. Based on the weight of each indicator in its respective indicator set and the standardized indicator values, the supply index and demand index of the target product category are determined. Finally, the supply-demand balance coefficient of the target product category can be determined based on the supply index and demand index. Therefore, in this embodiment, by incorporating heterogeneous indicators from multiple content carriers to construct supply-side and demand-side indicator sets, cross-carrier supply and demand data analysis is achieved. Combined with data standardization, the differences in dimensions and type conflicts of heterogeneous indicators are eliminated, providing a unified evaluation basis for multi-content carrier indicator data that were previously incomparable. This quantifies the platform's supply capacity and user demand across content carriers on a unified scale, automatically calculating the objective weights of each indicator to construct standardized supply and demand indices. This effectively assesses the platform's supply and demand health in the target category, significantly improving the comprehensiveness and consistency of supply and demand relationship assessments. Furthermore, this embodiment uses factor analysis instead of manual experience to determine indicator weights, avoiding biases caused by subjectively setting indicator weights. The indicator weights are determined by the standardized indicator values ​​in the supply-side and demand-side indicator sets, enabling dynamic responses to market supply and demand changes. Compared to the traditional fixed-weight model, this significantly improves the reliability, objectivity, and adaptability to real-world scenarios of supply and demand relationship assessment results.

[0073] based on Figure 2 In addition to the method described in the embodiments of this specification, some specific implementation schemes of the method are also provided, which will be described below.

[0074] In one optional implementation of this embodiment, the multiple content carriers may include at least two of the following: live streaming, product notes, and product cards.

[0075] Live streaming can refer to a real-time interactive video content format within e-commerce platforms. Merchants or live streamers showcase, explain, and promote products in real time via video streams, while users can watch, comment, ask questions, and purchase products simultaneously. As a content carrier, the core characteristics of live streaming are its real-time nature and highly interactive scenarios.

[0076] Product notes, in this context, refer to text, images, or short videos on e-commerce platforms that are associated with or used for product promotion. They are typically created by users, influencers, and merchants to share product experiences, reviews, tutorials, or recommendations. As a content carrier, the core characteristics of product notes are content retention, shareability, and the ability to generate interest-based product recommendations.

[0077] Product cards can refer to static or simply dynamic display pages that include information such as the product's main image, title, price, and basic attributes. As a content carrier, the core characteristics of product cards are direct information accessibility, search and recommendation entry points, and silent conversion scenarios.

[0078] In the embodiments of this specification, the constructed supply-side indicator set and demand-side indicator set both include indicators from multiple content carriers such as live streaming, product notes, and product cards. This overcomes the limitation of evaluation perspective caused by the reliance on a single content carrier data source in traditional methods, enabling the subsequently determined supply and demand indices to more realistically and completely reflect the supply and demand relationship of the target category in a complex platform ecosystem containing multiple content carriers.

[0079] In one optional implementation of this embodiment, the supply-side indicator set includes at least one of the following: product supply dimension indicators, content supply dimension indicators, merchant and streamer dimension indicators, and customer experience dimension indicators; the demand-side indicator set includes at least one of the following: search demand dimension indicators, carrier interaction dimension indicators, and purchase conversion demand dimension indicators.

[0080] Among them, commodity supply dimension indicators refer to a set of indicators used to quantitatively evaluate the richness, effectiveness, vitality, and price coverage of commodities within a target category. Commodity supply dimension indicators may include, but are not limited to: SPU richness, SPU effectiveness, proportion of new product SPUs, and price range coverage.

[0081] Among them, content supply dimension indicators refer to a set of metrics used to quantitatively evaluate a platform's ability to showcase and promote target product categories through content carriers such as live streaming, product notes, and product cards. Content supply dimension indicators may include, but are not limited to: product note dimension indicators, live streaming dimension indicators, and product card dimension indicators.

[0082] Among them, the merchant and livestreamer dimension indicators refer to a set of indicators used to quantitatively evaluate the scale, activity level, and sales performance of supply-side entities (merchants and livestreamers). Merchant and livestreamer dimension indicators may include, but are not limited to: merchant richness, merchant sales performance, livestreamer scale, and livestreamer sales performance.

[0083] Customer experience metrics refer to a set of indicators used to quantitatively evaluate negative feedback from users after purchasing and using target product categories. These metrics reflect problems and shortcomings in the existing supply chain's ability to meet user needs. Customer experience metrics may include, but are not limited to: return rates for standardized product units due to price issues, and return rates for standardized product units due to quality issues. A higher return rate indicates a worse customer experience.

[0084] Among them, search demand dimension metrics refer to a set of indicators used to quantitatively evaluate users' intention to demand products in a target category, as expressed through their proactive search behavior. Search demand dimension metrics may include, but are not limited to: search impression page views, search click page views, number of users searching, and number of searches.

[0085] Among them, the carrier interaction dimension metrics refer to a set of indicators used to quantitatively evaluate users' interest, engagement, and superficial conversion intentions during interactions with different content carriers (product notes, live streams, product cards). This dimension focuses on the interest and demand generated by "browsing and interaction." Carrier interaction dimension metrics may include, but are not limited to: the number of interactive users for each carrier, the interaction rate for each carrier, and the click-through rate for each carrier.

[0086] Purchase conversion demand metrics refer to a set of indicators used to quantitatively evaluate user behaviors such as completing a transaction or adding a product to their wish list to meet their needs. Purchase conversion demand metrics may include, but are not limited to: the number of products purchased across different platforms, and the number of users with purchase demand across different platforms.

[0087] In this embodiment, various supply-side indicators are categorized and statistically analyzed based on four dimensions: product supply, content supply, merchants and livestreamers, and customer experience. Similarly, various demand-side indicators are categorized and statistically analyzed based on three dimensions: search demand, carrier interaction, and purchase conversion demand. This ensures that the constructed indicator system can comprehensively and systematically reflect the essence of the supply and demand relationship.

[0088] In one optional implementation of this embodiment, the product supply dimension indicators include at least one of the following: standardized product unit richness, standardized product unit effectiveness, standardized product unit new product launch activity, and price range coverage; the content supply dimension indicators include at least one of the following: product note dimension indicators and live streaming dimension indicators; the product note dimension indicators include at least one of the following: number of product notes, product note exposure page views, number of unique visitors to product notes, percentage of products linked to product notes, product note page view click-through rate, product note unique visitor click-through rate, and percentage of product notes viewed for more than a preset duration; the live streaming dimension indicators include at least one of the following: live stream exposure page views, number of unique visitors to live stream, live stream page view click-through rate, live stream unique visitor click-through rate, and percentage of live stream viewers who watched for more than a preset duration; the merchant and streamer dimension indicators include at least one of the following: merchant richness, merchant sales performance, streamer scale, and streamer sales performance; the customer experience dimension indicators include at least one of the following: standardized product unit return rate due to price issues and standardized product unit return rate due to quality issues.

[0089] Specifically, the richness of standardized product units refers to the number of SKUs (Single Product Units) within a target category on the platform that have been exposed within a preset time period (e.g., the past 30 days). The effectiveness of standardized product units refers to the number of SKUs within a target category on the platform that have sales on the same day. The activity level of new standardized product units refers to the percentage of SKUs within a target category on the platform that have been exposed within a preset time period (e.g., the past 30 days) and are newly added within that preset time period.

[0090] Price band coverage refers to the dispersion or completeness of the price range distribution of goods within a target category. Price band coverage can be measured by calculating the entropy of the proportion of goods in different preset price ranges (e.g., low, medium, high), or by directly assessing whether key price points are covered. Price band coverage can be used to evaluate whether the supply can meet the needs of users at different levels of purchasing power.

[0091] The percentage of product notes attached to a product refers to the proportion of SPUs that are exposed in product notes on that day out of all SPUs in the target category. The percentage of product notes with browsing time exceeding the preset duration refers to the proportion of product notes in the target category that are viewed by users for more than the preset duration on that day. The preset duration can be set and adjusted according to actual needs; for example, it can be set to 10 seconds, 20 seconds, or 1 minute, without specific limitations.

[0092] The percentage of live streams with viewing time exceeding the preset duration refers to the proportion of live streams on that day that feature products of the target category and whose viewing time exceeds the preset duration. The preset duration can be set and adjusted according to actual needs; for example, it can be set to 30 seconds, 1 minute, or 10 minutes, without specific limitations.

[0093] Merchant richness refers to the number of merchants whose products belonging to the target category have been exposed within a preset time period (e.g., the past 30 days), after deduplication. Merchant sales momentum refers to the percentage of merchants whose products belonging to the target category have been exposed within a preset time period (e.g., the past 30 days) and have sales volume (sales volume of products in the target category).

[0094] Among these, "livestreamer scale" refers to the total number of livestreamers who have held livestream sessions to promote products in the target category within a preset time period (e.g., the past 30 days). "Livestreamer sales power" refers to the percentage of livestreamers who have generated sales (sales of products in the target category) among all livestreamers who have held livestream sessions to promote products in the target category within the preset time period (e.g., the past 30 days).

[0095] The return rate for standardized product units due to price issues refers to the percentage of SPUs (Sales Units) in a target product category that users returned due to "price issues" (such as feeling they paid too much, price fluctuations, etc.) within a preset time period (e.g., the past 30 days), out of the total SPUs sold in that category. This is a negative experience indicator that may reflect problems with the price competitiveness of the supply or the pricing strategy.

[0096] The standardized product unit return rate due to quality issues refers to the percentage of SPUs (Sales Units) in a target product category that were returned due to "quality issues" (such as defective products or products not matching the description) within a preset time period (e.g., the past 30 days), out of the total SPUs sold in that category. This is a negative experience indicator that reflects the quality and quality control level of the supplied goods.

[0097] In one optional implementation of this embodiment, the search demand dimension indicators include at least one of the following: search exposure page views, search click page views, number of search users, number of searches, and number of users who generate purchase intentions through search; the carrier interaction dimension indicators include at least one of the following: number of users who click on live streams, click-through rate of live stream products, number of users who interact with live streams, user interaction rate of live streams, number of users who click on product notes, click-through rate of product notes, number of users who interact with product notes, user interaction rate of product notes, and number of users who click on product cards; the purchase conversion demand dimension indicators include at least one of the following: number of products purchased through live streams, number of users with purchase intentions through live streams, number of products purchased through product notes, number of users with purchase intentions through product notes, number of products purchased through product cards, and number of users with purchase intentions through product cards.

[0098] Specifically, search impressions pageviews refer to the total number of times users view the search results page for relevant keywords in the target category after a user initiates a search within a statistical period (e.g., 24 hours). Search clicks pageviews refer to the total number of times users click on a specific result (such as a product, note, or live stream card) in the search results page within a statistical period (e.g., 24 hours). Search users refer to the number of unique users who have initiated search behavior related to the target category within a statistical period (e.g., 24 hours). Search counts refer to the total number of search queries related to the target category initiated by users within a statistical period (e.g., 24 hours).

[0099] The number of users who generate purchase intent through search refers to the number of unique users who initiated a search and subsequently expressed purchase intent within a statistical period (e.g., 24 hours). The following four behaviors can be considered as indicating that a user has generated purchase intent: 1. The user immediately purchased the product. 2. The user added the product to their shopping cart. 3. The user added the product to their wishlist. 4. The user saved the product to their favorites.

[0100] Among these, the number of users clicking on live streams, product notes, and product cards refers to the number of unique users who clicked on or entered the corresponding content platform (live stream, product notes, or product cards) after it gained exposure. The live stream product click-through rate (CTR) refers to the percentage of users who entered the live stream and subsequently clicked on the product links displayed or recommended within it. The product note CTR refers to the percentage of users who entered the note's details page and subsequently clicked on the product links displayed or recommended within it.

[0101] The number of interactive users refers to the number of unique users who have engaged in interactive behaviors such as liking, commenting, saving, forwarding, and posting comments within the corresponding content platform context. The user interaction rate refers to the proportion of users who interacted with the content platform out of the total number of users reached.

[0102] Among them, the number of products purchased through live streaming, the number of products purchased through product notes, and the number of products purchased through product cards refer to the quantity of products of the target category sold through the corresponding content platforms (live streaming, product notes, and product cards). The number of users with purchase intentions through live streaming, the number of users with purchase intentions through product notes, and the number of users with purchase intentions through product cards refer to the number of unique users who have purchased products of the target category or expressed purchase intentions for products of the target category through the corresponding content platforms (live streaming, product notes, and product cards).

[0103] See Figure 3 , Figure 3 This specification illustrates a flowchart of the indicator set acquisition process in a quantitative assessment method for supply and demand relationships, as provided in one embodiment. Figure 3 As shown, product supply dimension indicators, content supply dimension indicators, merchant and streamer dimension indicators, and customer experience dimension indicators can be obtained from the original data pool. Summarizing these indicators yields a supply-side indicator set. Similarly, search demand dimension indicators, platform interaction dimension indicators, and purchase conversion demand dimension indicators can be obtained from the original data pool. Summarizing these indicators yields a demand-side indicator set. In practical applications, each indicator in the indicator set can contain multiple sample values, allowing the determination of the corresponding feature matrix based on the data in the indicator set.

[0104] In one optional implementation of this embodiment, the step of standardizing the data of each indicator in the supply-side indicator set and the demand-side indicator set to obtain standardized indicator values ​​may specifically include: Determine the indicator categories of each indicator in the supply-side indicator set and the demand-side indicator set; the indicator categories include positive count categories, positive ratio categories, and negative categories; Based on the index categories of the aforementioned indicators, missing value imputation is performed on the indicators with missing values ​​to obtain the indicators after missing value imputation. Based on the index categories of the aforementioned indicators, data transformation processing is performed on the indicators after the missing values ​​are filled to obtain the transformed indicators. The transformed data is then subjected to standard normalization to obtain standardized index values.

[0105] In practical applications, positive indicators are those with higher exponential values, representing better business performance; negative indicators are those with higher exponential values, representing worse business performance. Negative indicators can include, but are not limited to, the return rate of standardized product units due to price issues and the return rate of standardized product units due to quality issues. Indicators classified as negative in both the supply-side and demand-side indicator sets have negative values. For example, the indicator "return rate of standardized product units due to quality issues" has a value of -0.1, meaning that within a preset time period, for the target product category, the proportion of SPUs (Sales Units) returned due to quality issues is 10% of the total SPUs sold in that category.

[0106] In practical applications, positive metrics can include positive count metrics and positive ratio metrics. Positive count metrics are characterized by non-negative integer values, a large numerical range, and typically a right-skewed (long-tailed) distribution. This right-skewed (long-tailed) distribution specifically refers to the data pattern in count-based business metrics (such as impressions and sales) where most samples have small values, but a few samples have extremely large values. Metrics categorized as positive counts may include, but are not limited to: standardized product unit richness, number of product notes, product note exposure page views, number of unique visitors to product notes, live stream exposure page views, number of unique visitors to live stream, merchant richness, streamer scale, search exposure page views, search click page views, number of search users, number of searches, number of users who generated purchase intent through search, number of users who clicked on live streams, number of users who interacted with live streams, number of users who clicked on product notes, number of users who interacted with product notes, number of users who clicked on product cards, number of products purchased during live streams, number of users with purchase intent during live streams, number of products purchased through product notes, number of users with purchase intent through product notes, number of products purchased through product cards, and number of users with purchase intent through product cards.

[0107] Among them, positive ratio metrics are characterized by values ​​typically falling within a bounded range of 0 to 1 (or 0% to 100%), with higher values ​​indicating better business performance. Metrics categorized as positive ratios may include, but are not limited to: standardized product unit effectiveness, standardized product unit new product launch activity, price range coverage, percentage of products linked to product notes, click-through rate (CTR) of product note page views, CTR of unique visitors to product notes, percentage of product notes viewed for more than a preset duration, CTR of live stream page views, CTR of unique visitors to live streams, percentage of live stream viewers viewed for more than a preset duration, merchant sales momentum, streamer sales momentum, live stream product user CTR, live stream user interaction rate, product note user CTR, and product note user interaction rate.

[0108] In practical applications, the category of an indicator can be determined based on the sign and dimensions of its value. Alternatively, for a specific indicator, its category can be pre-specified. Specifically, a correspondence between indicator names and indicator categories can be established, allowing the category of an indicator to be determined based on its name.

[0109] In the embodiments of this specification, missing value imputation refers to the operation of filling in null records in the indicator data with numerical values ​​according to the category to which the indicator belongs, using specific rules (such as filling with 0 or median). The purpose of missing value imputation is to eliminate the obstacles that null values ​​cause to subsequent statistical analysis (such as calculating the mean and standard deviation).

[0110] In practical applications, after imputing missing values ​​for indicators, data transformation processing can be performed on each indicator based on its category. This data transformation processing refers to applying specific mathematical functions to the indicator values ​​to further optimize the statistical characteristics of the data (e.g., making the data closer to a normal distribution) or to unify the evaluation direction of all indicators (so that increasing the value of all indicators indicates better performance).

[0111] In the embodiments of this specification, standard normalization refers to the process of applying the Z-Score standardization formula to the previously processed index data to convert the data into a standard normally distributed variable. The purpose of standard normalization is to completely eliminate the influence of dimensional differences and absolute value magnitudes between different indicators, so that all indicators are on the same scale.

[0112] Specifically, the formula for calculating the Z-Score is as follows:

[0113] Where Z represents the standardized index value; X represents the index value; This represents the mean; It represents the standard deviation.

[0114] In practical applications, an indicator can contain multiple sample values. For example, the live stream user interaction rate indicator might have sample value 1 as 0.2, sample value 2 as 0.4, and sample value 3 as 0.3. When performing standard normalization on these three sample values, the Z-score formula is used to first calculate the mean of the three sample values. =0.3, standard deviation ≈0.08165, the standardized index value of sample value 1 = (0.2-0.3) / 0.08165≈-1.225; the standardized index value of sample value 2 = (0.4-0.3) / 0.08165≈1.225; the standardized index value of sample value 3 = (0.3-0.3) / 0.08165=0.

[0115] In the embodiments described in this specification, missing values ​​in the indicators can be filled to improve the indicators and eliminate the impact of null values ​​on subsequent statistical analysis. By performing data transformation processing on the indicators after missing value filling, the statistical characteristics of the data can be further optimized, making the indicator data closer to a normal distribution. This also unifies the evaluation direction of each indicator. Finally, by performing standard normalization on each indicator, the dimensional differences between different indicators are eliminated, laying the foundation for subsequent data processing based on factor analysis.

[0116] In one optional implementation of this embodiment, the step of filling in missing values ​​for indicators with missing values ​​based on the indicator categories corresponding to each indicator may specifically include: For any of the aforementioned indicators, if the indicator category is the positive count category, then the missing value of the indicator is determined as a preset fill value. If the index category is the positive ratio category, then the missing values ​​of the index are determined as the median of each non-missing value sample in the index. If the index category is the negative category, then the missing value of the index is determined as the median of each non-missing value sample in the index.

[0117] The preset fill value can be set to 0 or a small value close to 0; there is no specific limitation on the specific value of the preset fill value. If the indicator category is positive counting, then the missing values ​​of the indicator can be determined as the preset fill values. For example, if the preset fill value is 0, and the indicator "Number of Product Notes" is a positive counting indicator, and if the three sample values ​​of the "Number of Product Notes" indicator are 22, missing, and 34, then the missing value of sample 2 can be determined as 0, and the three sample values ​​after missing value filling are 22, 0, and 34.

[0118] In practical applications, since an indicator can contain multiple sample values, it is usually not the case that all of the sample values ​​contained in an indicator are missing values. Therefore, if the indicator category is positive ratio or negative ratio, the missing value of the indicator can be determined as the median of each non-missing value sample in the indicator.

[0119] It's important to note that for indicators classified as positive counts, missing values ​​are filled with zeros. This strictly adheres to the objective business principle that "no record means no occurrence," ensuring that the filled data matrix accurately reflects the actual presence or absence of each product category across different channels or dimensions. This avoids inflating business scale due to incorrect filling (such as filling with the mean) and guarantees the accuracy of supply and demand "existence" assessments. Conversely, for indicators classified as positive ratios or negative ratios, missing values ​​are filled with the median, a robust strategy insensitive to outliers. Even with a few extremely high or low ratios in the valid sample, the median remains stable, ensuring that the filled values ​​are not distorted. This effectively prevents incorrect estimations of a large number of missing samples due to a few outliers, enhancing the robustness of the entire data processing flow.

[0120] In one optional implementation of this embodiment, the step of performing data transformation processing on the indicators after the missing values ​​are filled, based on the indicator categories of each indicator, to obtain the transformed indicators, may specifically include: For any of the indicators, if the indicator category is the positive count category, then take the logarithm of the indicator after filling in the missing values ​​to obtain the indicator after data transformation. If the index category is the positive ratio category, then the index after filling in the missing values ​​is determined as the index after data transformation. If the index category is the negative category, then the negative value of the index after filling in the missing values ​​is determined as the index after data transformation.

[0121] In practical applications, since positive count indicators typically exhibit a right-skewed (long-tailed) distribution, direct standardization can lead to excessively large extreme value weights. Therefore, a logarithmic transformation can be applied to positive count indicators to effectively compress the variable scale and make the transformed indicators closer to a normal distribution. The logarithm can refer to applying the natural logarithm to the value, i.e., calculating the corresponding value with a base of the constant e (approximately 2.71828), mathematically expressed as ln(x). Preferably, the indicator value can be incremented by 1 before taking the logarithm, i.e., ln(x+1), to avoid the mathematical error of ln(0).

[0122] In practical applications, since the values ​​of negative class indicators are usually negative, the negative values ​​of the indicators can be used as the indicators after data transformation. That is, for negative class indicators, data transformation is essentially to invert the negative values ​​of the indicators, so that the negative class indicators obtained after data transformation have positive values, thereby ensuring the correctness of subsequent standard normalization processing and factor loading direction.

[0123] In the embodiments of this specification, for the severely right-skewed distribution of positive count indicators, the ln(x+1) transformation is used, which can significantly compress the influence of extreme large values, making the transformed data closer to a normal distribution. This satisfies the basic assumptions of subsequent factor analysis methods regarding the data distribution shape, and ensures the accuracy and scientific nature of the estimation of statistical quantities such as weight calculation and variance contribution. Inverting negative indicators (such as return rates) can eliminate the potential for logical confusion in the comprehensive evaluation caused by contradictory indicator directions, ensuring the correctness of subsequent standard normalization processing and factor loading directions.

[0124] See Figure 4 , Figure 4 This specification illustrates a flowchart of data standardization processing in a quantitative assessment method for supply and demand relationships provided in one embodiment. Figure 4 As shown, the indicators can be categorized into three types based on their indicator class: positive technical indicators, positive ratio indicators, and negative indicators. Different missing value imputation strategies and data transformation strategies can be applied to indicators of different categories. After missing value imputation and data transformation, a unified standard normalization process can be performed.

[0125] See Figure 5 , Figure 5 This document illustrates a flowchart of factor analysis in a quantitative assessment method for supply and demand relationships, provided in one embodiment of this specification. Figure 5 As shown, before determining the weights of each indicator using factor analysis, an applicability test (KMO test and Bartlett's test for sphericity) can be performed first. Factor analysis can only be executed if the test passes; otherwise, the process ends to prevent wasting computational resources. After passing the test, the following steps of factor analysis can be performed: extracting common factors, rotating the factor loading matrix, and calculating the weights of each indicator within its respective indicator set.

[0126] In one optional implementation of this embodiment, determining the supply index and demand index of the target product category based on the weights and the values ​​of each standardized indicator may specifically include: Based on the standardized index values ​​of each indicator in the supply-side indicator set and the weights, the weighted values ​​of each indicator in the supply-side indicator set are determined. The supply index is obtained by summing the weighted values ​​of each indicator in the supply-side indicator set. Based on the standardized index values ​​of each indicator in the demand-side index set and the weights, the weighted values ​​of each indicator in the demand-side index set are determined. The demand index is obtained by summing the weighted values ​​of each indicator in the demand-side indicator set.

[0127] Specifically, the formula for calculating the supply index is:

[0128] in, This represents the supply index; n represents the number of concentrated supply-side indicators. This indicates the weight of supply-side indicator i; This represents the standardized value of supply-side indicator i.

[0129] Specifically, the formula for calculating the demand index is:

[0130] in, The index represents the demand index; m represents the number of demand-side indicators. This indicates the weight of demand-side indicator j; This represents the standardized value of demand-side indicator j.

[0131] In one optional implementation of this embodiment, determining the supply-demand balance coefficient of the target product category based on the supply index and the demand index may specifically include: The ratio of the demand index to the supply index is determined as the supply-demand balance coefficient for the target product category; Correspondingly, after determining the supply-demand balance coefficient of the target product category based on the supply index and the demand index, the process may further include: Based on the supply and demand balance coefficient, indication information is generated for adjusting the supply and demand relationship of the target product category.

[0132] Specifically, the formula for calculating the supply-demand balance coefficient is as follows:

[0133] in, This represents the supply and demand balance coefficient; Indicates the demand index; This represents the supply index.

[0134] In practical applications, after calculating the supply-demand balance coefficient of the target product category based on the demand index and the supply index, the supply and demand relationship of the target product category can be determined according to the value of the supply-demand balance coefficient. Specifically, when the supply-demand balance coefficient is greater than a preset threshold (for example, the preset threshold is 1), it indicates that market demand is greater than supply, and there is a supply gap; when the supply-demand balance coefficient is equal to the preset threshold, it means that supply and demand are in an ideal equilibrium state; when the supply-demand balance coefficient is less than the preset threshold, it means that supply is excessive and demand is relatively insufficient.

[0135] In practical applications, based on the Supply-Demand Balance (SDB) coefficient, indicators are generated to adjust the supply and demand relationship of target product categories. This ensures that resource allocation (such as traffic, budget, and manpower) is no longer based on vague feelings or egalitarianism, but on clear, quantifiable signals. For example, the algorithm team can prioritize optimizing the distribution efficiency of categories with high SDB, and the merchant recruitment team can precisely target and introduce high SDB categories. This ensures that limited platform resources are invested in categories with the most prominent supply-demand imbalances and the clearest return on investment, thereby maximizing the overall efficiency of resource utilization.

[0136] In one optional implementation of this embodiment, the step of generating indication information based on the supply-demand balance coefficient to indicate adjustments to the supply-demand relationship of the target product category may specifically include: Determine whether the supply-demand balance coefficient is greater than a preset threshold to obtain a first determination result; If the first judgment result indicates that the supply and demand balance coefficient is greater than the preset threshold, then a first indication information is generated to indicate an increase in the supply of the target product category; If the first judgment result indicates that the supply and demand balance coefficient is not greater than the preset threshold, then it is determined whether the supply and demand balance coefficient is less than the preset threshold to obtain the second judgment result; If the second judgment result indicates that the supply and demand balance coefficient is less than the preset threshold, then a second indication information is generated to indicate a reduction in the supply of the target product category; If the second judgment result indicates that the supply and demand balance coefficient is not less than the preset threshold, then a third indication information is generated to indicate the supply and demand balance of the target product category.

[0137] The preset threshold can refer to a critical value that is pre-set and stored in the system configuration. The specific value of the preset threshold can be set and adjusted based on business experience, historical data, or strategic goals. For example, it can be set to 1 (representing the ideal point of absolute supply and demand balance), or it can be set to other values ​​without specific limitations.

[0138] When the supply-demand balance coefficient is greater than a preset threshold, it indicates that the demand for the target product category exceeds the supply, resulting in a supply gap. Therefore, a first indication can be generated to instruct an increase in the supply of the target product category. When the supply-demand balance coefficient is less than a preset threshold, it indicates that the demand for the target product category is less than the supply, resulting in a surplus. Therefore, a second indication can be generated to instruct a decrease in the supply of the target product category.

[0139] Since the second step of judgment is only performed when the first judgment result indicates that the supply and demand balance coefficient is not greater than the preset threshold, if the second judgment result indicates that the supply and demand balance coefficient is not less than the preset threshold, it means that the supply and demand balance coefficient is neither greater than nor less than the preset threshold, that is, the supply and demand balance coefficient is equal to the preset threshold. In this case, third indication information can be generated to indicate the supply and demand balance of the target product category.

[0140] In the embodiments described in this specification, through a progressive logic of "first judgment" and "second judgment," the system can accurately categorize the supply-demand balance coefficient (SDB) of each product category into one of three distinct states: "supply shortage," "supply surplus," or "supply-demand balance," without omission or ambiguity. This automated diagnosis replaces the ambiguity and subjectivity of manual numerical interpretation, ensuring a high degree of consistency and repeatability in state judgment. Furthermore, the instruction information generated based on the supply-demand balance coefficient can efficiently and quickly issue accurate instructions to relevant personnel, assisting them in making accurate decisions.

[0141] See Figure 6 , Figure 6 This specification illustrates an overall flowchart of a quantitative assessment method for supply and demand relationships provided in one embodiment. Figure 6 As shown, after obtaining the supply-side and demand-side indicator sets for the target product category, the indicator categories of each indicator can be determined first. Then, missing value imputation and data transformation are performed based on the indicator categories, followed by standard normalization. After data processing, the weights of each indicator in its respective indicator set can be determined using factor analysis. The supply index, demand index, and supply-demand balance coefficient are then calculated sequentially. Finally, based on the supply-demand balance coefficient, indications for adjusting the supply-demand relationship of the target product category can be generated.

[0142] Corresponding to the above method embodiments, this specification also provides embodiments of a supply and demand relationship quantitative assessment device. Figure 7 A schematic diagram of a supply and demand relationship quantification assessment device according to one embodiment of this specification is shown. Figure 7 As shown, the device may include: The acquisition module 702 is configured to acquire a supply-side indicator set and a demand-side indicator set for the target product category; wherein, both the supply-side indicator set and the demand-side indicator set contain indicators of multiple content carriers, and at least two of the indicators of the multiple content carriers are heterogeneous to each other. The data processing module 704 is configured to perform data standardization processing on each indicator in the supply-side indicator set and the demand-side indicator set to obtain the values ​​of each standardized indicator. The first determining module 706 is configured to determine the weight of each indicator in its respective indicator set based on factor analysis, according to the standardized indicator values ​​of the supply-side indicator set and the demand-side indicator set. The second determining module 708 is configured to determine the supply index and demand index of the target product category based on the weights and the values ​​of each standardized indicator. The third determining module 710 is configured to determine the supply-demand balance coefficient of the target product category based on the supply index and the demand index.

[0143] In one optional embodiment, the multiple content carriers include at least two of the following: live streaming, product notes, and product cards.

[0144] In one optional embodiment, the supply-side indicator set includes at least one of the following: product supply dimension indicators, content supply dimension indicators, merchant and streamer dimension indicators, and customer experience dimension indicators; the demand-side indicator set includes at least one of the following: search demand dimension indicators, carrier interaction dimension indicators, and purchase conversion demand dimension indicators.

[0145] In an optional embodiment, the product supply dimension indicators include at least one of the following: standardized product unit richness, standardized product unit effectiveness, standardized product unit new product launch activity, and price range coverage; the content supply dimension indicators include at least one of the following: product note dimension indicators and live streaming dimension indicators; the product note dimension indicators include at least one of the following: number of product notes, product note exposure page views, number of unique visitors to product notes, percentage of products linked to product notes, product note page view click-through rate, product note unique visitor click-through rate, and percentage of product notes viewed for more than a preset duration; the live streaming dimension indicators include at least one of the following: live stream exposure page views, number of unique visitors to live stream, live stream page view click-through rate, live stream unique visitor click-through rate, and percentage of live stream viewers who watched for more than a preset duration; the merchant and streamer dimension indicators include at least one of the following: merchant richness, merchant sales performance, streamer scale, and streamer sales performance; the customer experience dimension indicators include at least one of the following: standardized product unit return rate due to price issues and standardized product unit return rate due to quality issues.

[0146] In one optional embodiment, the search demand dimension metrics include at least one of search exposure page views, search click page views, number of search users, number of searches, and number of users who generate purchase intentions through search; the carrier interaction dimension metrics include at least one of live stream click users, live stream product click-through rate, live stream interactive users, live stream user interaction rate, product note click users, product note product click-through rate, product note interactive users, product note user interaction rate, and product card click users; the purchase conversion demand dimension metrics include at least one of live stream product purchases, live stream purchase demand users, product note purchases, product note purchase demand users, product card purchases, and product card purchase demand users.

[0147] In an optional embodiment, the data processing module 704 is further configured to: Determine the indicator categories of each indicator in the supply-side indicator set and the demand-side indicator set; the indicator categories include positive count categories, positive ratio categories, and negative categories; Based on the index categories of the aforementioned indicators, missing value imputation is performed on the indicators with missing values ​​to obtain the indicators after missing value imputation. Based on the index categories of the aforementioned indicators, data transformation processing is performed on the indicators after the missing values ​​are filled to obtain the transformed indicators. The transformed data is then subjected to standard normalization to obtain standardized index values.

[0148] Optionally, the step of filling in missing values ​​for indicators with missing values ​​based on the indicator categories corresponding to each indicator may specifically include: For any of the aforementioned indicators, if the indicator category is the positive count category, then the missing value of the indicator is determined as a preset fill value. If the index category is the positive ratio category, then the missing values ​​of the index are determined as the median of each non-missing value sample in the index. If the index category is the negative category, then the missing value of the index is determined as the median of each non-missing value sample in the index.

[0149] Optionally, the step of performing data transformation processing on the indicators after filling in the missing values, based on the indicator categories of each indicator, to obtain the transformed indicators, may specifically include: For any of the indicators, if the indicator category is the positive count category, then take the logarithm of the indicator after filling in the missing values ​​to obtain the indicator after data transformation. If the index category is the positive ratio category, then the index after filling in the missing values ​​is determined as the index after data transformation. If the index category is the negative category, then the negative value of the index after filling in the missing values ​​is determined as the index after data transformation.

[0150] In an optional embodiment, the second determining module 708 is further configured to: Based on the standardized index values ​​of each indicator in the supply-side indicator set and the weights, the weighted values ​​of each indicator in the supply-side indicator set are determined. The supply index is obtained by summing the weighted values ​​of each indicator in the supply-side indicator set. Based on the standardized index values ​​of each indicator in the demand-side index set and the weights, the weighted values ​​of each indicator in the demand-side index set are determined. The demand index is obtained by summing the weighted values ​​of each indicator in the demand-side indicator set.

[0151] In an optional embodiment, the third determining module 710 is further configured to: The ratio of the demand index to the supply index is determined as the supply-demand balance coefficient for the target product category; After determining the supply-demand balance coefficient of the target product category based on the supply index and the demand index, the method further includes: Based on the supply and demand balance coefficient, indication information is generated for adjusting the supply and demand relationship of the target product category.

[0152] Optionally, the step of generating indication information based on the supply-demand balance coefficient to indicate adjustments to the supply-demand relationship of the target product category may specifically include: Determine whether the supply-demand balance coefficient is greater than a preset threshold to obtain a first determination result; If the first judgment result indicates that the supply and demand balance coefficient is greater than the preset threshold, then a first indication information is generated to indicate an increase in the supply of the target product category; If the first judgment result indicates that the supply and demand balance coefficient is not greater than the preset threshold, then it is determined whether the supply and demand balance coefficient is less than the preset threshold to obtain the second judgment result; If the second judgment result indicates that the supply and demand balance coefficient is less than the preset threshold, then a second indication information is generated to indicate a reduction in the supply of the target product category; If the second judgment result indicates that the supply and demand balance coefficient is not less than the preset threshold, then a third indication information is generated to indicate the supply and demand balance of the target product category.

[0153] The supply and demand relationship quantitative assessment device provided in the embodiments of this specification, after obtaining the supply-side indicator set and demand-side indicator set of the target product category, can perform data standardization processing on each indicator in the supply-side indicator set and demand-side indicator set to obtain the standardized indicator values. Both the supply-side indicator set and the demand-side indicator set contain indicators with multiple content carriers, and at least two of the indicators with multiple content carriers are heterogeneous. Then, based on factor analysis, the weight of each indicator in its respective indicator set is determined according to the standardized indicator values ​​in the supply-side indicator set and the demand-side indicator set. Based on the weight of each indicator in its respective indicator set and the standardized indicator values, the supply index and demand index of the target product category are determined. Finally, the supply and demand balance coefficient of the target product category can be determined based on the supply index and the demand index. Therefore, in this embodiment, by incorporating heterogeneous indicators from multiple content carriers to construct supply-side and demand-side indicator sets, cross-carrier supply and demand data analysis is achieved. Combined with data standardization, the differences in dimensions and type conflicts of heterogeneous indicators are eliminated, providing a unified evaluation basis for multi-content carrier indicator data that were previously incomparable. This quantifies the platform's supply capacity and user demand across content carriers on a unified scale, automatically calculating the objective weights of each indicator to construct standardized supply and demand indices. This effectively assesses the platform's supply and demand health in the target category, significantly improving the comprehensiveness and consistency of supply and demand relationship assessments. Furthermore, this embodiment uses factor analysis instead of manual experience to determine indicator weights, avoiding biases caused by subjectively setting indicator weights. The indicator weights are determined by the standardized indicator values ​​in the supply-side and demand-side indicator sets, enabling dynamic responses to market supply and demand changes. Compared to the traditional fixed-weight model, this significantly improves the reliability, objectivity, and adaptability to real-world scenarios of supply and demand relationship assessment results.

[0154] The above is a schematic scheme of a supply and demand relationship quantitative assessment device according to this embodiment. It should be noted that the technical solution of this supply and demand relationship quantitative assessment device and the technical solution of the supply and demand relationship quantitative assessment method described above belong to the same concept. For details not described in detail in the technical solution of the supply and demand relationship quantitative assessment device, please refer to the description of the technical solution of the supply and demand relationship quantitative assessment method described above.

[0155] Figure 8A structural block diagram of a computing device 800 according to one embodiment of this specification is shown. The components of the computing device 800 include, but are not limited to, a memory 810 and a processor 820. The processor 820 is connected to the memory 810 via a bus 830, and a database 850 is used to store data.

[0156] The computing device 800 also includes an access device 840, which enables the computing device 800 to communicate via one or more networks 860. Examples of these networks include Public Switched Telephone Network (PSTN), Local Area Network (LAN), Wide Area Network (WAN), Personal Area Network (PAN), or combinations of communication networks such as the Internet. The access device 840 may include one or more of any type of wired or wireless network interface (e.g., a network interface card (NIC)), such as an IEEE 802.11 Wireless Local Area Network (WLAN) wireless interface, a Wi-MAX (Worldwide Interoperability for Microwave Access) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, or a Near Field Communication (NFC) interface.

[0157] In one embodiment of this specification, the above-described components of the computing device 800 and Figure 8 Other components, not shown, can also be connected to each other, for example, via a bus. It should be understood that... Figure 8 The block diagram of the computing device shown is for illustrative purposes only and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0158] The computing device 800 can be any type of stationary or mobile computing device, including mobile computers or mobile computing devices (e.g., tablet computers, personal digital assistants, laptop computers, notebook computers, netbooks, etc.), mobile phones (e.g., smartphones), wearable computing devices (e.g., smartwatches, smart glasses, etc.) or other types of mobile devices, or stationary computing devices such as desktop computers or personal computers (PCs). The computing device 800 can also be a mobile or stationary server.

[0159] The processor 820 is used to execute the following computer-executable instructions, which, when executed by the processor, implement the steps of the above-mentioned quantitative assessment method for supply and demand relationship.

[0160] The above is an illustrative scheme of a computing device according to this embodiment. It should be noted that the technical solution of this computing device and the technical solution of the above-described quantitative assessment method for supply and demand belong to the same concept. For details not described in detail in the technical solution of the computing device, please refer to the description of the technical solution of the above-described quantitative assessment method for supply and demand.

[0161] An embodiment of this specification also provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the above-described quantitative assessment method for supply and demand relationships.

[0162] The above is an illustrative scheme of a computer-readable storage medium according to this embodiment. It should be noted that the technical solution of this storage medium belongs to the same concept as the technical solution of the above-described quantitative assessment method for supply and demand. For details not described in detail in the technical solution of the storage medium, please refer to the description of the technical solution of the above-described quantitative assessment method for supply and demand.

[0163] An embodiment of this specification also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described quantitative assessment method for supply and demand relationships.

[0164] The above is an illustrative scheme of a computer program product according to this embodiment. It should be noted that the technical solution of this computer program product and the technical solution of the above-described quantitative assessment method for supply and demand belong to the same concept. For details not described in detail in the technical solution of the computer program product, please refer to the description of the technical solution of the above-described quantitative assessment method for supply and demand.

[0165] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0166] The computer instructions include computer program code, which may be in the form of source code, object code, executable file, or certain intermediate forms. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium may be appropriately added or removed according to the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media may not include electrical carrier signals and telecommunication signals.

[0167] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments in this specification are not limited to the described order of actions, because according to the embodiments in this specification, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the embodiments in this specification.

[0168] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0169] The preferred embodiments disclosed above are merely illustrative of this specification. Optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the embodiments described in this specification. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the embodiments, thereby enabling those skilled in the art to better understand and utilize this specification.

Claims

1. A method for quantitatively assessing supply and demand relationships, characterized in that, The method includes: Obtain the supply-side indicator set and the demand-side indicator set of the target product category; wherein, both the supply-side indicator set and the demand-side indicator set contain indicators of multiple content carriers, and at least two of the indicators of the multiple content carriers are heterogeneous to each other. The indicators in the supply-side indicator set and the demand-side indicator set are standardized to obtain the values ​​of each standardized indicator. Based on factor analysis, the weight of each indicator in its respective indicator set is determined according to the standardized indicator values ​​in the supply-side indicator set and the demand-side indicator set. Based on the weights and the values ​​of each standardized indicator, the supply index and demand index of the target product category are determined. The supply-demand balance coefficient of the target product category is determined based on the supply index and the demand index.

2. The quantitative assessment method for supply and demand relationship according to claim 1, characterized in that, The various content formats include at least two of the following: live streaming, product notes, and product cards.

3. The quantitative assessment method for supply and demand relationship according to claim 1, characterized in that, The supply-side indicators include at least one of the following: product supply dimension indicators, content supply dimension indicators, merchant and streamer dimension indicators, and customer experience dimension indicators; the demand-side indicators include at least one of the following: search demand dimension indicators, carrier interaction dimension indicators, and purchase conversion demand dimension indicators.

4. The quantitative assessment method for supply and demand relationship according to claim 3, characterized in that, The product supply dimension indicators include at least one of the following: richness of standardized product units, effectiveness of standardized product units, activity of new product launches in standardized product units, and price range coverage; the content supply dimension indicators include at least one of the following: product note dimension indicators and live streaming dimension indicators; the product note dimension indicators include at least one of the following: number of product notes, page views of product note exposure, number of unique visitors to product note exposure, percentage of products linked to product notes, click-through rate of product note page views, click-through rate of unique visitors to product notes, and percentage of product notes viewed for more than a preset duration; the live streaming dimension indicators include at least one of the following: page views of live stream exposure, number of unique visitors to live stream exposure, click-through rate of live stream page views, click-through rate of unique visitors to live stream, and percentage of live stream viewers who watched for more than a preset duration; the merchant and streamer dimension indicators include at least one of the following: merchant richness, merchant sales performance, streamer scale, and streamer sales performance; the customer experience dimension indicators include at least one of the following: return rate of standardized product units due to price issues and return rate of standardized product units due to quality issues.

5. The quantitative assessment method for supply and demand relationship according to claim 3, characterized in that, The search demand dimension metrics include at least one of the following: search exposure page views, search click page views, number of search users, number of searches, and number of users who generate purchase intentions through search; the platform interaction dimension metrics include at least one of the following: number of users who click on live streams, click-through rate of live stream products, number of users who interact with live streams, user interaction rate of live streams, number of users who click on product notes, click-through rate of product notes, number of users who interact with product notes, user interaction rate of product notes, and number of users who click on product cards; the purchase conversion demand dimension metrics include at least one of the following: number of products purchased through live streams, number of users with purchase intentions through live streams, number of products purchased through product notes, number of users with purchase intentions through product notes, and number of users with purchase intentions through product cards.

6. The quantitative assessment method for supply and demand relationship according to claim 1, characterized in that, The process of standardizing the data of each indicator in the supply-side indicator set and the demand-side indicator set to obtain standardized indicator values ​​includes: Determine the indicator categories of each indicator in the supply-side indicator set and the demand-side indicator set; the indicator categories include positive count categories, positive ratio categories, and negative categories; Based on the index categories of the aforementioned indicators, missing value imputation is performed on the indicators with missing values ​​to obtain the indicators after missing value imputation. Based on the index categories of each index, data transformation processing is performed on each index after the missing values ​​are filled to obtain the indexes after data transformation. The transformed data is then subjected to standard normalization to obtain standardized index values.

7. The quantitative assessment method for supply and demand relationship according to claim 6, characterized in that, The step of filling in missing values ​​for indicators with missing values ​​based on the indicator categories corresponding to each indicator includes: For any of the indicators, if the indicator category is the positive count category, then the missing value of the indicator is determined as the preset fill value. If the index category is the positive ratio category, then the missing values ​​of the index are determined as the median of each non-missing value sample in the index. If the index category is the negative category, then the missing value of the index is determined as the median of each non-missing value sample in the index.

8. The quantitative assessment method for supply and demand relationship according to claim 6, characterized in that, Based on the indicator categories of each indicator, the missing value-filled indicators are subjected to data transformation processing to obtain the transformed indicators, including: For any of the indicators, if the indicator category is the positive count category, then take the logarithm of the indicator after filling in the missing values ​​to obtain the indicator after data transformation. If the index category is the positive ratio category, then the index after filling in the missing values ​​is determined as the index after data transformation. If the index category is the negative category, then the negative value of the index after filling in the missing values ​​is determined as the index after data transformation.

9. The quantitative assessment method for supply and demand relationship according to claim 1, characterized in that, The step of determining the supply index and demand index of the target product category based on the weights and the values ​​of each standardized indicator includes: Based on the standardized index values ​​of each indicator in the supply-side indicator set and the weights, the weighted values ​​of each indicator in the supply-side indicator set are determined. The supply index is obtained by summing the weighted values ​​of each indicator in the supply-side indicator set. Based on the standardized index values ​​of each indicator in the demand-side index set and the weights, the weighted values ​​of each indicator in the demand-side index set are determined. The demand index is obtained by summing the weighted values ​​of each indicator in the demand-side indicator set.

10. The quantitative assessment method for supply and demand relationship according to claim 1, characterized in that, Determining the supply-demand balance coefficient of the target product category based on the supply index and the demand index includes: The ratio of the demand index to the supply index is determined as the supply-demand balance coefficient for the target product category; After determining the supply-demand balance coefficient of the target product category based on the supply index and the demand index, the method further includes: Based on the supply and demand balance coefficient, indication information is generated for adjusting the supply and demand relationship of the target product category.

11. The quantitative assessment method for supply and demand relationship according to claim 10, characterized in that, The step of generating indication information based on the supply-demand balance coefficient to indicate adjustments to the supply-demand relationship of the target product category includes: Determine whether the supply-demand balance coefficient is greater than a preset threshold to obtain a first determination result; If the first judgment result indicates that the supply and demand balance coefficient is greater than the preset threshold, then a first indication information is generated to indicate an increase in the supply of the target product category; If the first judgment result indicates that the supply and demand balance coefficient is not greater than the preset threshold, then it is determined whether the supply and demand balance coefficient is less than the preset threshold to obtain the second judgment result; If the second judgment result indicates that the supply and demand balance coefficient is less than the preset threshold, then a second indication information is generated to indicate a reduction in the supply of the target product category; If the second judgment result indicates that the supply and demand balance coefficient is not less than the preset threshold, then a third indication information is generated to indicate the supply and demand balance of the target product category.

12. A device for quantitatively assessing supply and demand relationships, characterized in that, The device includes: The acquisition module is configured to acquire a supply-side indicator set and a demand-side indicator set for the target product category; wherein, both the supply-side indicator set and the demand-side indicator set contain indicators of multiple content carriers, and at least two of the indicators of the multiple content carriers are heterogeneous to each other. The data processing module is configured to perform data standardization processing on each indicator in the supply-side indicator set and the demand-side indicator set to obtain the values ​​of each standardized indicator. The first determining module is configured to determine the weight of each indicator in its respective indicator set based on factor analysis, according to the standardized indicator values ​​of the supply-side indicator set and the demand-side indicator set. The second determining module is configured to determine the supply index and demand index of the target product category based on the weights and the values ​​of each standardized indicator. The third determining module is configured to determine the supply-demand balance coefficient of the target product category based on the supply index and the demand index.

13. A computing device, characterized in that, include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the method according to any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that, It stores computer-executable instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 11.

15. A computer program product, characterized in that, It includes a computer program or instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 11.

Images