Internet of Things industrial development evaluation method and device

Abstract

The invention discloses an Internet of Things industrial development evaluation method and device, and the method comprises the steps: obtaining a data source related to a development plan, carrying out the collection and arrangement of the data source through a text mining technology, and generating a standardized text database; performing word segmentation and stop word removal processing on the original text in the standardized text database, and establishing a text feature set through word frequency distribution statistics and feature extraction to obtain a text feature result; based on a text feature result, performing index design by utilizing an evaluation framework of a development goal, a support tool and a supporting measure dimension, and constructing a multi-input-output table; performing data processing through a PMC index calculation model by utilizing evaluation indexes in the multi-input-output table to obtain a PMC index evaluation result; and based on a PMC index evaluation result, performing evaluation through a consistency analysis model to form a development improvement suggestion. The problems that in the prior art, an evaluation method is high in subjectivity and low in evaluation precision, and a systematic evaluation framework is lacked are effectively solved.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a method and device for evaluating the development of the Internet of Things industry. Background Art

[0002] As one of the core technologies of the digital economy, the Internet of Things (IoT) is a crucial enabler for the digital transformation of industries. Since the 1990s, the development of IoT technology has evolved from perception and recognition to network interconnection and then to intelligent analysis.

[0003] Currently, fuzzy comprehensive evaluation and analytic hierarchy process are commonly used to evaluate industrial development. For example, some researchers use bibliometrics to analyze development and evolution characteristics, while others use instrumental theory to explore structural issues.

[0004] The more mature approach currently is to use text mining methods to identify features and extract information from texts, interpret the deep semantic connotations of texts through technical means such as machine learning and natural language processing, and then conduct quantitative evaluation of the effects.

[0005] However, existing evaluation methods suffer from strong subjectivity and low accuracy. Furthermore, quantitative evaluation research on the development of my country's IoT industry is largely limited to the macro level, with relatively little regional assessment. Furthermore, existing research lacks a systematic evaluation framework, making it difficult to provide scientific decision-making support for optimization. Summary of the Invention

[0006] The purpose of the present invention is to provide a method and device for evaluating the development of the Internet of Things industry, so as to solve technical problems in the existing technology such as the strong subjectivity of evaluation methods, low evaluation accuracy, and lack of a systematic evaluation framework.

[0007] To achieve the above objectives, the present invention provides a method for evaluating the development of the Internet of Things industry, comprising: Acquire data sources related to development planning, collect and organize the data sources using text mining technology, and generate a standardized text database; Perform word segmentation and stop word removal on the original text in the standardized text database, establish a text feature set by statistical word frequency distribution and feature extraction, and obtain text feature results; Based on the text feature results, the evaluation framework of development goals, support tools and supporting measures is used to design indicators and construct a multi-input-output table; Using the evaluation indicators in the multi-input-output table, data processing is performed through a preset PMC index calculation model to obtain a PMC index evaluation result; Based on the PMC index evaluation results, a system evaluation is conducted through a preset consistency analysis model to form development and improvement suggestions.

[0008] Preferably, the data source is collected and sorted using text mining technology to generate a standardized text database, specifically including: Using keyword indexing technology to screen the data source, and generating a multi-source data set through automatic collection and processing; Performing format recognition and coding standardization processing on the multi-source data set to establish a standardized data index; Divide the data into different levels based on the normalized data index and generate a multi-level data structure through classification and sorting; Designing text identifiers using the multi-level data structure and generating a unique identifier sequence through automatic encoding; Data integration is performed based on the unique identification sequence and the multi-level data structure to generate the standardized text database.

[0009] Preferably, the word segmentation and stop word removal processing of the original text in the standardized text database specifically includes: Obtaining the original text of the standardized text database, building a professional dictionary library using domain knowledge, and forming a basic word segmentation vocabulary; Perform text segmentation processing based on the basic word segmentation vocabulary to obtain an initial word segmentation result; Performing grammatical structure analysis on the initial word segmentation results and obtaining a basic phrase set through part-of-speech tagging; Using the basic phrase set to perform synonym analysis and word form unification, and obtain standard phrases through regularization processing; Stop words are identified and filtered based on the standard phrases to obtain a valid phrase set.

[0010] Preferably, the method is to design indicators based on the text feature results using an evaluation framework of development goals, support tools, and supporting measures to construct a multi-input-output table, including: Obtaining a feature vector of the text feature result, and using the evaluation framework to design an indicator to form an evaluation indicator framework; Design quantitative rules based on the evaluation indicator framework and establish indicator calculation standards through numerical mapping; The qualitative characteristics are numerically processed using the index calculation standard, and quantitative data are obtained through conversion rules; Designing a matrix structure based on the quantitative data to obtain a multidimensional data matrix through dimensional organization; The multidimensional data matrix is ​​used to establish an input-output mapping relationship, and a multi-input-output table is obtained through modeling.

[0011] Preferably, the method of using the evaluation indicators in the multi-input-output table to process data using a preset PMC index calculation model to obtain a PMC index evaluation result specifically includes: The coefficient analysis is performed using the text type and the level of the issuing unit, and the text strength coefficient α value is obtained through hierarchical evaluation calculation; Analyze the implementation period data based on the text strength coefficient α value, and obtain the implementation period parameter β value through time sequence rule processing; The implementation period parameter β value is used to classify the indicators of the multi-input-output table, and the input and output variable sets are obtained through the variable identification algorithm; Performing data standardization based on the input and output variable sets, and obtaining a standardized indicator matrix through dimensional conversion; The standardized indicator matrix is ​​combined with the coefficient parameters to obtain the PMC index evaluation result containing the PMC index value through a preset PMC calculation model.

[0012] Preferably, after obtaining the PMC index evaluation result including the PMC index value by using the standardized indicator matrix in combination with the coefficient parameter through a preset PMC calculation model, the method further comprises: Using the PMC index evaluation results to design a coordinate system, and constructing a three-dimensional display space through dimensional mapping; Performing data positioning based on the three-dimensional display space and obtaining a scattered point distribution set through spatiotemporal mapping; Performing surface fitting using the scattered point distribution set, and obtaining an initial surface using a cubic spline interpolation algorithm; Optimizing the smoothness of the initial surface and obtaining an optimized surface model by adjusting parameters; The optimized surface model is used for visualization processing, and a PMC surface graph is obtained through color mapping and node labeling.

[0013] Preferably, based on the PMC index evaluation results, a system evaluation is performed through a preset consistency analysis model to form development and improvement suggestions, including: Based on the PMC index evaluation result, adjacent text data is extracted using the PMC surface graph, and the change rate feature is obtained by comparing the indicators; Perform hierarchical analysis based on the change rate characteristics and obtain a consistency score through matching calculation; Perform difference analysis using the consistency score and obtain difference feature vectors by index comparison; Identify influencing factors based on the difference feature vectors and obtain an influence degree matrix through correlation analysis; The influence degree matrix is ​​used to perform feature extraction, and key influencing factors are obtained through attribution analysis.

[0014] Preferably, based on the PMC index evaluation results, a system evaluation is performed through a preset consistency analysis model to form development and improvement suggestions, including: Obtain the PMC index evaluation results and establish an adaptive mind map framework; Use the adaptive mind mapping framework to break down complex problems into interrelated sub-problems; Constructing a directed acyclic graph structure, prioritizing the subproblems and analyzing the subproblems according to the priority order, and obtaining analysis results of the subproblems; Integrate the analysis results of the sub-problems to form the development improvement suggestions; Continuously update the data of the PMC index evaluation results and optimize the problem decomposition and analysis strategy.

[0015] Preferably, based on the PMC index evaluation results, a system evaluation is performed through a preset consistency analysis model to form development and improvement suggestions, including: Build a policy language agent system and establish a latent space model for policy representation; A strategy combination is sampled from the latent space model to obtain a strategy combination, which is applied to the evaluation and optimization process; Establish a multi-level strategy feedback mechanism to evaluate the execution effect of the strategy combination and obtain the strategy execution evaluation results; updating the latent space distribution of the latent space model according to the strategy execution evaluation result to generate an optimization strategy; Comprehensively optimize evaluation strategies, analysis strategies, and implementation strategies to improve overall performance.

[0016] Preferably, the method further comprises: Obtaining evaluation indicators in the multi-input-output table and establishing a mixed linear regression model; A technique combining tensor decomposition and spectral methods is used to ensure global convergence; The key parameters of the PMC index are estimated using a gradient-based stochastic approximation method; Use Bayesian nonparametric mixture models to handle nonlinear relationships in data; A parameter learning algorithm is constructed to ensure the model convergence of the hybrid linear regression model, and the PMC index evaluation result containing the PMC index value is obtained.

[0017] The present invention also provides an Internet of Things industry development evaluation device, comprising: A data acquisition module is used to acquire data sources related to the development plan, collect and organize the data sources using text mining technology, and generate a standardized text database; A text processing module is used to perform word segmentation and stop word removal on the original text in the standardized text database, establish a text feature set by statistical word frequency distribution and feature extraction, and obtain text feature results; An indicator construction module is used to design indicators based on the text feature results and use an evaluation framework of development goals, supporting tools, and supporting measures to construct a multi-input-output table; An index calculation module, configured to use the evaluation indicators in the multi-input-output table to perform data processing through a preset PMC index calculation model to obtain a PMC index evaluation result; The optimization suggestion module is used to perform a system evaluation based on the PMC index evaluation results through a preset consistency analysis model to form development improvement suggestions.

[0018] The beneficial effects of the present invention are: 1. This paper proposes a development quantitative evaluation method based on the combination of text mining and the PMC index model, which can effectively reduce the subjectivity of the evaluation process and improve the evaluation accuracy; 2. This invention innovatively incorporates both macro and regional levels into the evaluation system, expanding the evaluation dimensions and making development assessment more comprehensive and systematic; 3. This invention designs a multi-dimensional evaluation index system that includes development goals, supporting tools, supporting measures, implementation effects, etc., providing a more scientific evaluation framework; 4. The present invention establishes a dynamic feedback mechanism to achieve continuous optimization of evaluation and ensure the timeliness and accuracy of evaluation results; 5. This paper proposes a visualization method based on PMC surface to intuitively display the consistency level, making it easier for decision makers to understand and use the evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 A flowchart of the method for evaluating the development of the Internet of Things industry provided by an embodiment of the present invention; Figure 2 Flowchart of PMC index calculation and consistency evaluation provided by an embodiment of the present invention; Figure 3This is a structural block diagram of the Internet of Things industry development evaluation device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0022] See Figure 1 , Figure 1 This is a flow chart of a method for evaluating the development of the Internet of Things industry provided by an embodiment of the present invention. The method includes: Step 101: Acquire data sources related to development planning, collect and organize the data sources using text mining technology, and generate a standardized text database.

[0023] Specifically, this step first identifies the data sources required for IoT industry development evaluation, including various development planning documents, industry support documents, technical standards and specifications, industry reports, and other textual materials. Automated crawler technology is then used to collect relevant text from public websites, professional databases, and industry platforms. The collected textual materials are formatted and structured. During this process, the text sources are verified and quality-checked, and duplicate, invalid, or low-quality text is eliminated. Next, the text is classified and coded according to the professional knowledge system in the IoT field, and a multidimensional index system is established, including dimensions such as publication time, technical field, and development stage. Finally, the processed text is stored in a structured database to construct an IoT industry text corpus, providing a standardized data foundation for subsequent analysis. This step ensures that the data sources for the evaluation process are comprehensive, authoritative, and timely, and is a fundamental component of the entire evaluation method. In practice, different data collection strategies will be implemented for different IoT sub-sectors (such as smart manufacturing, smart agriculture, and smart home) to ensure the professionalism and targeted nature of the text database.

[0024] Step 102: performing word segmentation and stop word removal processing on the original text in the standardized text database, establishing a text feature set by statistical word frequency distribution and feature extraction, and obtaining a text feature result.

[0025] In this step, first read the original documents in the text database and perform Chinese word segmentation in combination with the professional vocabulary in the Internet of Things field to accurately identify the professional terms and compound concepts related to the Internet of Things. For special words in the Internet of Things field, such as "RFID", "M2M", "edge computing", etc., special rules are used for identification and annotation to ensure the accuracy of word segmentation. Then, perform word tagging on the word segmentation results to identify different types of words such as nouns, verbs, and adjectives, and filter out words that have no substantial meaning for analysis, such as common function words like "de", "shi", "he", etc., according to the predefined stop word list. Next, the system calculates the frequency of occurrence and distribution characteristics of each word in the text, and uses the TF-IDF algorithm to evaluate the importance of each word in the text of the Internet of Things industry development, and identify the keyword vocabulary that can reflect the industry characteristics. In addition, semantic features of the text are also extracted, including multi-dimensional features such as theme features, development trends, and technical relevance. Finally, these word frequency features and semantic features are integrated into a structured feature vector to form the text feature result, providing data support for the subsequent index design and evaluation system construction. In practical applications, the Internet of Things professional dictionary will be updated regularly to adapt to the rapidly evolving technical terms in this field and improve the accuracy of word segmentation and feature extraction.

[0026] Step 103: Based on the text feature result, use the evaluation framework of the development goal, support tools, and supporting measures dimensions to design indicators and construct a multi-input-output table.

[0027] In this step, based on the foregoing text feature analysis results, a multi-dimensional index system for the evaluation of the Internet of Things industry development is constructed. First, extract the key indicators reflecting the development goals of the Internet of Things industry from the text features, including dimensions such as the growth goal of the industry scale, the level of technological innovation, and the degree of application popularization; secondly, extract the indicators related to support tools, covering aspects such as the support intensity, the scale of capital investment, and the market access conditions; thirdly, extract the indicators related to supporting measures, including elements such as the talent cultivation system, the completeness of standard construction, and the construction of the ecological environment. Quantify these extracted qualitative indicators, and use a combination of expert scoring and fuzzy comprehensive evaluation to convert the text description into measurable values. Then, design the multi-dimensional index matrix structure, and classify and organize each index into input indicators and output indicators according to the upstream and downstream relationship of the Internet of Things industrial chain and the characteristics of the development stage. Finally, establish the logical mapping relationship between the indicators and construct a complete multi-input-output table, which not only reflects the internal structure and operation mechanism of the Internet of Things industry development, but also provides a scientific data basis for the calculation of the PMC index. In practice, this index system will be dynamically adjusted according to the evolution of Internet of Things technologies and the expansion of applications to ensure the forward-looking and adaptability of the evaluation indicators.

[0028] Step 104: Using the evaluation indicators in the multi-input-output table, data is processed using a preset PMC index calculation model to obtain a PMC index evaluation result.

[0029] This step is the core of IoT industry development evaluation. Based on the evaluation indicators in the multi-input-output table, a systematic evaluation is conducted using the PMC index calculation model. First, the system determines the text strength coefficient α based on the document type and issuing unit level, reflecting the guiding strength and implementation rigidity of different documents in IoT industry development. Then, the implementation deadlines and timing of each document are analyzed to determine the implementation cycle parameter β, reflecting the time span and execution cycle of different development documents. Next, the indicator data in the multi-input-output table is standardized to eliminate dimensional differences and make them comparable. The system then calculates the indicator matching and strategy consistency between documents at different levels, generating a consistency coefficient γ to reflect the degree of coordination in IoT industry development at different levels. Finally, the system calculates the PMC index value for each evaluation object based on the PMC index calculation formula, comprehensively considering the text strength, implementation cycle, and consistency level. As a comprehensive evaluation indicator, the PMC index reflects both the overall level of the IoT industry and the characteristics and interrelationships of different development stages, providing a quantitative basis for industry development decision-making. The system also generates a PMC surface plot to visually display the index's trend over time and level, enabling decision-makers to quickly grasp the development status. During the specific implementation process, the PMC index model can set different parameter weights for different segments of the Internet of Things to reflect the development characteristics and focus of each segment.

[0030] The PMC index calculation model in this invention is a mathematical model based on a comprehensive evaluation of multiple factors. It is called the Public Text Modeling Consistency Analysis Framework. It is used to quantify the consistency level of IoT industry development. The core concept of this model is to weight and integrate three key factors: text strength, implementation cycle, and hierarchical coordination to form a comprehensive evaluation index. Its mathematical expression is: PMC = α·P + β·M + γ·C P represents the power factor, which measures the guiding strength and implementation rigidity of the development plan document; M represents the maturity factor, reflecting the time span and execution cycle of the development plan; and C represents the coordination factor, indicating the level of consistency between development plans at different levels. α, β, and γ are the weight coefficients of these three factors, satisfying the equation α + β + γ = 1.

[0031] In the specific implementation, the calculation of P factor adopts the analytic hierarchy process (AHP), which determines the relative importance weight of each text according to the text type, publishing unit level and implementation rigidity. The system first constructs the judgment matrix A, and the matrix element a ij It represents the importance ratio of text i to text j, and then calculates the maximum eigenvalue λmax of the matrix and the corresponding eigenvector w. After normalization, the eigenvector is used to obtain the weight coefficient of each text, thereby calculating the P factor value.

[0032] The M factor is calculated using the time-weighted average method, taking into account the release time, planned implementation time, and coverage period of each document. The specific calculation formula is:

[0033] Among them, w i Represents the weight coefficient of the i-th text, t i The time parameter representing the text is calculated based on the recency of the text's publication time and the length of the planned implementation period.

[0034] The C-factor is calculated using a text similarity algorithm to assess the degree of content consistency across development plan documents at different levels. The system uses the TF-IDF vector space model to convert text into high-dimensional vector representations and then calculates the cosine similarity between these vectors as a consistency indicator. For multiple documents, a similarity matrix is ​​constructed, and the average or minimum similarity is calculated as the overall consistency assessment.

[0035] Ultimately, the PMC index is derived through a weighted combination of the three factors mentioned above, ranging from 0 to 1. Values ​​closer to 1 indicate a higher level of consistency in IoT industry development. To improve the model's adaptability, the system dynamically adjusts the weighting coefficients of the three factors based on the characteristics of different regions and development stages, ensuring the accuracy and relevance of the evaluation results. Furthermore, the PMC index calculation takes time continuity into account, introducing historical data smoothing to reduce index fluctuations and reflect the stable trends of industrial development.

[0036] Step 105: Based on the PMC index evaluation results, a system evaluation is performed through a preset consistency analysis model to form development and improvement suggestions.

[0037] In the final step, the PMC index evaluation results are thoroughly analyzed and systematically evaluated to identify existing challenges and areas for improvement in IoT industry development. First, the system extracts key data points from the PMC surface graph, analyzes the timing and trends of index fluctuations, and identifies key turning points in the industry. Next, the system compares PMC indices at different levels to assess the degree of coordination between macro-planning and regional implementation, identifying inconsistencies across levels. Next, through variance analysis, the system pinpoints the specific indicators and influencing factors that lead to inconsistencies, generating a problem diagnosis report. Finally, using correlation analysis, the system maps PMC index fluctuations to external factors, identifying the key factors most significantly impacting IoT industry development. Finally, based on these analysis results, the system generates targeted development improvement recommendations across dimensions such as industry goal alignment, resource allocation optimization, support tool selection, and implementation path design. These recommendations, informed by an objective analysis of historical data and a deep understanding of the dynamics of IoT industry development, provide a scientific basis for the formulation and adjustment of industry plans. The system also provides a dynamic feedback mechanism to continuously update evaluation results and optimization recommendations as new data is continuously input, ensuring the timeliness and accuracy of the evaluation. In actual application, the system can generate differentiated development suggestions based on the industrial base and development stage of different regions, thereby improving the pertinence and operability of the suggestions.

[0038] The proposed consistency analysis model is a mathematical model specifically designed to assess the coordination of IoT industry development. It analyzes the PMC index results from both vertical and horizontal perspectives. Based on structural equation modeling (SEM) and multidimensional scaling (MDS) techniques, the model can identify the degree of coordination and the causes of differences across different levels and regions.

[0039] The longitudinal consistency analysis employed structural equation modeling to construct a hierarchical transmission path model for the development of the IoT industry. This model incorporates three latent variables at the macro, meso, and micro levels, each with multiple observed variables corresponding to specific components of the PMC index. Model parameters were calculated using maximum likelihood estimation, and the path coefficients and goodness-of-fit between the various levels were assessed. The path coefficients reflect the strength of influence between the upper and lower levels, while the goodness-of-fit indicates the degree to which the overall model matches the actual data. Based on these parameters, the system identifies weaknesses in hierarchical transmission, such as the failure of macro-level planning to be effectively transmitted to regional implementation or the disconnect between regional planning and corporate actions.

[0040] Horizontal consistency analysis uses a multidimensional scaling algorithm and cluster analysis method to map the PMC index and its components of different regions into a two-dimensional space, visually displaying the similarities and differences between regions. The system first constructs the inter-region dissimilarity matrix D, where the matrix elements dij represent the weighted Euclidean distance between region i and region j along each dimension of the PMC index. Then, through an iterative optimization algorithm, it searches for a set of points P in two-dimensional space such that the distances between the points maintain the original dissimilarity relationship as much as possible, namely:

[0041] where ||p i -p j || represents the Euclidean distance between points i and j in two-dimensional space. The optimized two-dimensional point set forms a regional consistency scatter plot, where closer distances between points indicate greater regional development consistency. The system then applies a K-means or hierarchical clustering algorithm to group these points, identifying clusters of regions with similar development characteristics.

[0042] The model also incorporates a consistency decomposition algorithm that breaks down overall consistency differences into specific indicator dimensions, pinpointing the key factors that contribute to inconsistency. This algorithm uses analysis of variance (ANOVA) and relative importance analysis to calculate the contribution of each indicator to the overall variance, thereby identifying the dominant factors influencing consistency levels.

[0043] The consistency analysis model not only provides static assessment results but also conducts dynamic evolution analysis. By constructing a time series model, it can predict future trends in consistency levels. The system uses an autoregressive integrated moving average (ARIMA) model or a long short-term memory (LSTM) model to process historical data on the PMC index, capturing its cyclical fluctuations and long-term trends, providing early warning and guidance for future development.

[0044] The above steps are described in detail below with reference to specific embodiments.

[0045] The steps of constructing the Internet of Things industry development text corpus provided in the embodiment of the present invention include: Step 201: Filter the data source using keyword indexing technology, and generate a multi-source data set through automatic collection and processing.

[0046] Specifically, in this example, we first used keywords such as "Internet of Things" to collect IoT-related documents from 2006 to 2023 through legal and regulatory databases, official portals at all levels, and other channels. Specifically, we adhered to the three principles of authority, relevance, and representativeness: authority refers to selecting officially released documents; relevance refers to documents with direct relevance to the development of the IoT industry; and representativeness refers to documents with strong guiding significance.

[0047] Step 202: Perform format recognition and code standardization processing on the multi-source data set to establish a standardized data index.

[0048] Specifically, the collected texts were initially screened, eliminating documents such as letters and approvals, as well as those whose content did not specifically address the development of the IoT. Furthermore, the text formatting needed to be standardized, including removing special characters and using a unified encoding format, to prepare for subsequent analysis.

[0049] Step 203: Divide the data into hierarchies based on the normalized data index, and generate a multi-level data structure through classification and organization.

[0050] Specifically, the screened texts are classified and organized at the macro and regional levels, and a multi-level text structure is established based on the text type (such as planning, implementation, etc.).

[0051] Step 204: Design a text identifier using the multi-level data structure, and generate a unique identifier sequence through automatic encoding.

[0052] Step 205: Perform data integration based on the unique identification sequence and the multi-level data structure to generate the standardized text database.

[0053] Specifically, each text is assigned a unique identification code to facilitate subsequent tracing and analysis.

[0054] The steps of using software to perform text preprocessing and feature extraction provided in the embodiment of the present invention include: Step 301: obtaining the original text of the standardized text database, building a professional dictionary library using domain knowledge, and forming a basic word segmentation vocabulary library.

[0055] Step 302: Perform text segmentation processing based on the basic word segmentation vocabulary to obtain an initial word segmentation result.

[0056] Specifically, in this embodiment, ROST CM6 software was used to perform Chinese word segmentation and part-of-speech tagging on the text. Special attention was paid to processing specialized terms and compound words in the field of IoT to ensure accurate segmentation. Furthermore, a specialized dictionary for the field of IoT was established to assist in the segmentation process.

[0057] Step 303: Perform grammatical structure analysis on the initial word segmentation result, and obtain a basic phrase set through part-of-speech tagging.

[0058] Step 304: Perform synonym analysis and word form unification using the basic phrase set, and obtain standard phrases through regularization.

[0059] Step 305: performing stop word identification and filtering based on the standard phrases to obtain a valid phrase set.

[0060] Specifically, the segmentation results are filtered for stop words to remove function words and modal particles that are of no real significance to the analysis. Furthermore, data cleaning is required to process synonyms and antonyms, unify word forms, and improve the accuracy of subsequent analysis.

[0061] Next, we use the Term Frequency-Inverse Document Frequency (TF-IDF) algorithm to extract key feature words from the text and calculate the weight of each feature word. Through word frequency statistical analysis, we identify high-frequency words and key concepts in the text and reveal key areas of focus.

[0062] The steps of constructing a multi-input-output evaluation system of a PMC index model provided by an embodiment of the present invention include: Step 401: Acquire a feature vector of the text feature result, use the evaluation framework to design indicators, and form an evaluation indicator framework.

[0063] Specifically, based on the results of the feature word analysis, an evaluation index system was constructed from the dimensions of development goals, supporting tools, supporting measures, and implementation effects. Development goals include industrial scale and technological innovation; supporting tools include fiscal and tax support and market access; and supporting measures include talent development and standardization.

[0064] Step 402: Design quantitative rules based on the evaluation index framework and establish index calculation standards through numerical mapping.

[0065] Specifically, we designed quantitative standards for each evaluation indicator, using a combination of quantitative and qualitative methods. Quantitative indicators were directly measured using numerical values, while qualitative indicators were quantified using expert scoring to ensure objectivity.

[0066] Step 403: quantify the qualitative features using the indicator calculation standard and obtain quantitative data through conversion rules.

[0067] Step 404: Design a matrix structure based on the quantified data, and obtain a multi-dimensional data matrix through dimensional organization.

[0068] Step 405: Establish an input-output mapping relationship using the multidimensional data matrix, and obtain a multi-input-output table through modeling.

[0069] Specifically, the quantified evaluation index data is organized into a multi-input-output matrix. The rows of the matrix represent different texts, the columns represent the evaluation indicators, and the matrix elements represent the corresponding quantitative values.

[0070] See Figure 2 , Figure 2This is a flowchart of the PMC index calculation and consistency evaluation provided by an embodiment of the present invention. The process includes: Step 501: Perform coefficient analysis using text type and issuing unit level information, and calculate the text strength coefficient α value through hierarchical evaluation.

[0071] Specifically, the key parameters in the PMC index model need to be clarified. Specifically, for each document, its text strength coefficient α needs to be identified. This coefficient reflects the degree of compulsivity of the text and is determined based on the document type and the level of the issuing department. For example, the α value for important documents is 1.0, the α value for general documents is 0.8, and the α value for regular documents is 0.6.

[0072] Step 502: Analyze the implementation period data based on the text strength coefficient α value, and obtain the implementation period parameter β value through time sequence rule processing.

[0073] Specifically, it is necessary to determine the implementation period parameter β, which represents the expected time it takes for the text to produce its effects. This parameter can be determined by analyzing the implementation period explicitly stated in the text or by referencing the implementation experience of similar texts. For example, a β value of 5 years is typically used for planning documents, while a β value of 2-3 years is used for implementation details.

[0074] Step 503: using the implementation cycle parameter β value to classify the indicators of the multiple input-output table, and obtaining the input and output variable sets through the variable identification algorithm.

[0075] Specifically, the evaluation indicators are categorized into input-based and output-based variables. Input-based variables include the level of support, the scale of capital investment, and the completeness of supporting measures. Larger values ​​for these variables indicate greater investment. Output-based variables include the growth rate of industry scale, the degree of improvement in technological innovation capabilities, and the number of market entities cultivated. Larger values ​​for these variables indicate greater results.

[0076] Step 504: performing data standardization processing based on the input and output variable sets, and obtaining a standardized indicator matrix through dimensional conversion.

[0077] Specifically, the indicator data in the multi-input-output matrix are standardized to eliminate the dimensional differences between different indicators. The standardization uses the range method to uniformly convert the indicator values ​​into the interval [0,1].

[0078] Step 505: Utilizing the standardized indicator matrix in combination with the coefficient parameters, a preset PMC calculation model is used to obtain the PMC index evaluation result including the PMC index value.

[0079] Specifically, the consistency level is calculated. For texts in a time series, the rate of change of each indicator value between adjacent texts is calculated to assess the continuity of the adjustment. At the same time, considering the synergy between texts at different levels, the matching degree of macro and regional texts in each indicator dimension is calculated.

[0080] Finally, according to the PMC index calculation formula:

[0081] where γ i is the text consistency coefficient, n is the number of texts, and the PMC index value of each text is obtained. The closer the PMC index value is to 1, the higher the consistency level.

[0082] Next, the calculation results of the PMC index can be visualized: Step 506: Design a coordinate system using the PMC index evaluation result, and construct a three-dimensional display space through dimensional mapping.

[0083] To visualize the PMC index results calculated above, we first need to construct a coordinate system suitable for displaying the spatial distribution characteristics of the PMC index. Specifically, the system designs a three-dimensional spatial coordinate system. The X-axis typically represents the time dimension, which is scaled according to the time series of IoT industry development (e.g., quarterly or annually), from the earliest evaluation time point to the latest. The Y-axis represents the regional or hierarchical dimension, which can be divided according to different regional types (e.g., developed regions, emerging regions, basic regions) or management levels (e.g., macro level, regional level, enterprise level). The Z-axis represents the PMC index value, which typically ranges from 0 to 1, with values ​​closer to 1 indicating higher consistency. When constructing the coordinate system, the system considers the distribution characteristics of the data and appropriately sets the axial scale and scale to ensure clarity and interpretability of the final display. The system also sets appropriate reference planes and grid lines in the coordinate system to help viewers better understand the positional relationships of the data in three-dimensional space. After the coordinate system design is completed, the system will convert the PMC index evaluation results into spatial position information in the coordinate system through dimensional mapping rules, preparing for subsequent data visualization processing.

[0084] Step 507: performing data positioning based on the three-dimensional display space, and obtaining a scattered point distribution set through spatiotemporal mapping.

[0085] Specifically, a three-dimensional coordinate system is established, where the X-axis represents the time dimension, the Y-axis represents the text level dimension (macro, regional), and the Z-axis represents the PMC index value. The calculated PMC index values ​​are plotted in the coordinate system according to the time series and text level.

[0086] After the coordinate system is constructed, the system maps the PMC index data for each evaluation object to a specific location in three-dimensional space. For each PMC index evaluation record, the system determines the X coordinate value based on its time attribute, the Y coordinate value based on its region or level attribute, and the Z coordinate value based on its PMC index value, thus forming a data point in three-dimensional space. Once all evaluation records are mapped, the system generates a three-dimensional scattered point set consisting of multiple data points. The distribution of these scattered points intuitively reflects the changes in the consistency level of IoT industry development across time and space. The system then performs a preliminary analysis of these scattered points, identifying dense and sparse areas of scattered point distribution and possible anomalies and clustering. To enhance the visualization of the scattered point distribution set, the system also assigns different colors to the scattered points based on different PMC index value ranges. For example, the high index value range (0.8-1.0) may be represented by red, the medium index value range (0.5-0.8) by yellow, and the low index value range (0-0.5) by blue. Furthermore, the system can further enhance the data's expressiveness by adjusting visual attributes such as the size and shape of scattered points. This visualization of scattered point distribution allows decision makers to quickly grasp the overall trend and local characteristics of IoT industry development consistency.

[0087] Step 508: Surface fitting is performed using the scattered point distribution set, and an initial surface is obtained by using a cubic spline interpolation algorithm.

[0088] Although the scattered point distribution set can intuitively display the discrete distribution state of the PMC index, it is difficult to directly reflect the continuous change trend of the data. In order to more comprehensively display the change law of the PMC index, the system needs to construct a continuous and smooth surface model based on the scattered point distribution set. In this step, the system uses the cubic spline interpolation algorithm to perform surface fitting on the scattered data. The core idea of ​​the cubic spline interpolation algorithm is to construct a cubic polynomial function between adjacent data points so that the fitting surface not only passes through all the original data points, but also has good smoothness and continuity. In specific implementation, the system first divides the XY plane of the three-dimensional space into a regular grid, and then applies cubic spline interpolation to the data points in each grid to calculate the Z value of each point in the grid. For grid areas with missing data, the system uses the interpolation results of adjacent grids for reasonable inference. By processing all grid areas one by one, the system finally obtains a continuous surface covering the entire XY plane. The advantage of this method is that it can accurately retain the characteristics of the original data while filling in the blank areas of the data and presenting the overall trend of data changes. After the initial fitting is completed, the system will calculate the error between the fitted surface and the original data points, evaluate the fitting quality, and provide a reference for subsequent surface optimization.

[0089] Step 509: Optimizing the smoothness of the initial surface and obtaining an optimized surface model by adjusting parameters.

[0090] Specifically, a cubic spline interpolation method is used to fit the discrete PMC index values ​​to generate a continuous PMC surface. By adjusting the smoothness parameter of the surface, it is ensured that the surface can reflect both the overall trend of consistency and the local characteristics.

[0091] The initially fitted surface may have localized unevenness or overfitting, requiring further optimization. In this step, the system optimizes the smoothness of the initial surface by adjusting key parameters of the cubic spline interpolation algorithm. The main optimization methods include: first, adjusting the tension parameter to control the surface's "tightness." Higher tension values ​​result in a flatter surface, while lower tension values ​​result in a surface that is closer to the original data points. Second, a smoothing factor is introduced to reduce local surface fluctuations. A larger smoothing factor results in a smoother surface, but this may increase deviation from the original data points. Third, the system applies a local weighting strategy to address potential outliers, minimizing their impact on the overall surface morphology. Based on the actual needs of IoT industry development evaluation, the system seeks the optimal balance between maintaining surface morphological characteristics and ensuring fitting accuracy. The optimization process is iterative, calculating the fitting error and smoothness index after each iteration. The optimization process ends when both indicators reach a satisfactory level or when the preset number of iterations has been reached. The final optimized surface model can not only accurately reflect the overall distribution and changing trend of the PMC index, but also has good visual effects and interpretability, providing decision makers with clearer data insights.

[0092] The optimized surface model in this paper is a mathematical model for PMC index visualization and trend analysis. It uses an improved Bezier surface algorithm and adaptive smoothing technology to achieve high-precision and aesthetically pleasing surface fitting. The key innovation of this model lies in combining traditional cubic spline interpolation with machine learning regularization methods, overcoming the overfitting and fluctuation problems that standard interpolation algorithms are prone to when dealing with irregular data distributions.

[0093] The mathematical expression of the optimized surface model is:

[0094] Among them, B i,n (x) and B j,m (y) are the nth and mth order B-spline basis functions in the x-direction and y-direction respectively, P ij is the control point grid that controls the shape and characteristics of the surface. The objective function of the model optimization process is:

[0095] The first term is the data fitting error term, which indicates the degree of deviation of the surface from the original data points; the second term R(S) is the regularization term, which is used to control the smoothness and complexity of the surface; λ is the smoothing parameter, which controls the balance between fitting accuracy and smoothness.

[0096] In its implementation, an adaptive mesh subdivision strategy is employed, dynamically adjusting the density of the control point grid based on the density and gradient of the data distribution. In areas with dense and drastically varying data, the system increases the number of control points to improve fitting accuracy; in areas with sparse or gently varying data, the system reduces the number of control points to maintain surface smoothness. This adaptive strategy significantly improves fitting results, especially for the uneven data distribution common in the development of the IoT industry.

[0097] To further optimize the surface quality, the system introduces a surface optimization algorithm based on a physical model, which treats the surface as a thin plate with elastic potential energy and achieves natural smoothness by minimizing bending energy. Its potential energy function is:

[0098] The calculus of variations solves this minimization problem, yielding the optimal surface morphology. Furthermore, the model incorporates a surface feature preservation algorithm that maintains overall smoothness while retaining important features such as peaks and valleys in the data, thus avoiding information loss caused by over-smoothing.

[0099] The optimized surface model not only allows for static visualization but also supports dynamic evolution analysis. By constructing a three-dimensional surface model of a time series, the system generates an animation of the spatiotemporal evolution of the PMC index for IoT industry development, visually demonstrating the speed, direction, and acceleration of the index's changes. This dynamic visualization approach can help decision-makers better understand the historical trajectory and future trends of industrial development, providing an intuitive reference for strategic planning.

[0100] Step 510: Perform visualization processing using the optimized surface model, and obtain a PMC surface graph through color mapping and node labeling.

[0101] Specifically, by setting different color ranges, the PMC index's numerical distribution can be intuitively displayed. For example, the PMC index can be represented by red, yellow, and blue, respectively, according to the high, medium, and low ranges, allowing decision makers to quickly identify the strength of the consistency distribution. Furthermore, events at key time points are annotated in the surface chart to help analyze the impact of adjustments on the consistency level.

[0102] After optimizing the surface model, the system performs final visualization, generating an intuitive and easy-to-read PMC surface plot. First, the system applies a color mapping scheme to the optimized surface, assigning colors based on the Z value (i.e., PMC index) of each point on the surface. Common color mapping schemes include heatmaps, where high-index areas are displayed in red, medium-index areas in yellow, and low-index areas in blue. This gradient of colors visually illustrates the changing trends of the index. Next, the system annotates key nodes and characteristic areas on the surface, such as peaks and valleys, stable regions, and rapidly changing regions of the PMC index. Textual descriptions are provided to help decision makers understand the significance of these characteristic points. Clear scale labels and dimensional descriptions are also added to the coordinate axes to ensure the readability of the surface plot. To enhance the three-dimensional effect, appropriate lighting models and shading are applied to enhance the three-dimensionality of the surface. Furthermore, the system provides interactive browsing capabilities, allowing decision makers to observe the PMC surface from different angles and distances, and even slice through it to view the PMC index distribution at specific points in time or in specific areas. The resulting PMC surface graph not only has a pleasant visual aesthetic but, more importantly, can intuitively demonstrate the spatiotemporal distribution characteristics and changing patterns of IoT industry development consistency, providing strong data support for decision makers. In practical applications, this surface graph can be regularly updated to reflect the latest evaluation results, forming a dynamic evaluation mechanism.

[0103] Finally, development improvement suggestions are formed based on the PMC index evaluation results. The specific implementation process is as follows: Step 601: Based on the PMC index evaluation result, adjacent text data is extracted using the PMC surface graph, and a change rate feature is obtained by comparing the indicators.

[0104] Specifically, by comparing the PMC index values ​​at adjacent time points, the rate of change is calculated, and the time periods when the index rises or falls rapidly are identified as the key analysis objects.

[0105] Step 602: Perform hierarchical analysis based on the change rate feature and obtain a consistency score through matching calculation.

[0106] Specifically, the PMC index matching of macro- and regional-level texts is calculated to assess the degree of synergy between texts at different levels. The matching degree can be quantified by calculating the correlation coefficient or the degree of difference between the PMC indices of the two levels.

[0107] Step 603: Perform difference analysis using the consistency score, and obtain a difference feature vector by index comparison.

[0108] Specifically, for regions and time periods with low matching, we further analyze the specific indicators and factors that cause inconsistencies. The difference values ​​of each indicator are combined into a difference feature vector to reflect the specific manifestations of the inconsistency.

[0109] Step 604: Identify influencing factors based on the difference feature vectors, and obtain an influence degree matrix through correlation analysis.

[0110] Specifically, a correlation model between the PMC index and various influencing factors can be established using methods such as multiple regression analysis or structural equation modeling. Through model analysis, the degree of influence of each influencing factor on the PMC index is calculated to form an influence degree matrix.

[0111] Step 605: Extract features using the influence matrix and obtain key influencing factors through attribution analysis.

[0112] Specifically, based on the impact matrix, we identify the key factors that have the greatest impact on the PMC index. These key factors are the focus of optimization recommendations.

[0113] Based on the PMC index evaluation results, the PMC surface plot is used to extract adjacent text data, and rate-of-change features are derived through index comparison. This step first requires a detailed interpretation of the PMC surface plot to identify key feature points and regions within the surface. The system automatically extracts adjacent data points along the timeline of the PMC index and calculates the rate of change and trend of the index. Specifically, the system focuses on time periods where the PMC index shows significant increases or decreases, calculates the rate of change of the index during these time periods, and expresses it as a percentage or slope. Time periods with significant rates of change (e.g., those exceeding ±10%) are marked as critical observation periods for focused analysis. The system also compares the consistency of PMC index trends across different levels to determine whether macroeconomic development directions and regional implementation are synchronized. In this way, the system can capture important turning points and trends in the development of the IoT industry, providing a foundation for subsequent analysis. In practical applications, the system will combine background knowledge of industry development to provide a preliminary interpretation of these rate-of-change features, such as identifying cyclical variations that may be related to technological innovation cycles or industrial investment cycles.

[0114] Based on the aforementioned rate of change characteristics, a hierarchical analysis is performed, and a consistency score is calculated through matching. In this step, the system further analyzes the level of development consistency between different levels. In specific implementation, the system calculates the matching degree of the PMC index between the macro and regional levels. This matching degree can be quantified using a correlation coefficient or a difference index. The correlation coefficient is calculated using the Pearson correlation coefficient method, with a value range of [-1, 1]. A value closer to 1 indicates a more consistent development trend between the two levels. The difference index calculates the absolute or relative difference between the PMC indices of the two levels. A value closer to 0 indicates a closer development level between the two levels. The system calculates matching indicators for different time windows (such as the near-term, medium-term, and long-term) to generate consistency scores across multiple timescales. Furthermore, the system analyzes the consistency level between different regions to assess the balance and coordination of regional development. These consistency scores provide a quantitative basis for identifying coordination issues in IoT industry development, helping decision makers accurately grasp the coordination status between different levels. During implementation, the system will set thresholds for consistency scores. For example, a score below 0.5 will be considered low consistency, requiring specific attention and improvement.

[0115] The consistency scores are used to perform a variance analysis, generating a variance feature vector through index comparison. After identifying regions or time periods with low consistency scores, the system further analyzes the specific causes of the discrepancies. In this step, the system compares and analyzes specific evaluation indicators at different levels, calculates the variance of each indicator, and generates a variance feature vector. Each element of the variance feature vector represents the degree of variance in a specific indicator, providing a visual indication of the areas where significant differences exist across different levels. For example, the system may identify significant disparities between macro and regional levels in technological innovation investment, or inconsistent progress in standards development across different regions. The system then ranks the variance feature vectors and identifies the indicators with the greatest contribution (e.g., those with a variance greater than twice the average). These indicators are often the key factors contributing to a decline in overall consistency. This allows the system to identify key indicators requiring focus from a wide range of indicators, providing precise guidance for subsequent improvement recommendations. In practical applications, the system also incorporates expertise in IoT industry development to provide reasonable explanations for these differences, such as possible differences due to varying regional industrial bases or development stages.

[0116] Based on the differential eigenvectors, influencing factors are identified and an influence matrix is ​​generated through correlation analysis. After identifying key differential indicators, the system further analyzes the underlying causes of these differences. In this step, the system constructs a correlation model between the PMC index and each influencing factor. Using statistical methods such as multiple regression analysis, path analysis, or structural equation modeling, the system explores the path and extent of each factor's impact on the PMC index. The system first establishes a set of candidate influencing factors based on two dimensions: internal factors of IoT development (such as technological innovation capabilities, industry chain sophistication, and talent pool availability) and external factors (such as changes in market demand, capital investment, and the competitive environment). Then, through data mining and model calculations, the system determines the correlation between these factors and changes in the PMC index. The influence coefficient of each factor is calculated to form an influence matrix. The rows of this matrix represent different influencing factors, the columns represent different evaluation indicators or regions, and the matrix elements represent the degree of impact of the corresponding influencing factor on a specific indicator or region. By analyzing the influence matrix, the system can identify the key factors that most significantly influence the consistency of IoT industry development, providing a scientific basis for developing targeted improvement measures. During the specific implementation process, the system will adopt multiple analysis methods for cross-validation to improve the reliability of the results. For example, correlation analysis and causal inference techniques will be used simultaneously to ensure that the identified impact relationships have practical significance rather than just statistical correlations.

[0117] The influence matrix is ​​used to extract features, and key influencing factors are identified through attribution analysis. This step deepens and refines the factor analysis. Based on the influence matrix, the system further extracts and summarizes key influencing factors. In specific implementation, the system first performs factor cluster analysis, grouping similar influencing factors into the same category to reduce problem complexity. The system then calculates the combined influence of each factor category and identifies the key factors with the greatest impact on the PMC index. Based on these key factors, the system conducts attribution analysis to further explore the specific mechanisms and pathways through which each factor influences the development of the IoT industry. The attribution analysis is conducted from both qualitative and quantitative perspectives. The quantitative analysis primarily calculates the contribution rate and sensitivity coefficient of each factor to changes in the PMC index. The qualitative analysis combines theoretical knowledge and practical experience in IoT industry development to explain the mechanisms of each factor's influence. Through this comprehensive analysis, the system generates a structured list of key influencing factors, including a ranking of each factor's importance, an explanation of its influencing mechanisms, and an analysis of related pathways. These key influencing factors serve as the direct basis for formulating subsequent development and improvement recommendations, ensuring their relevance and effectiveness. In actual applications, the system will generate differentiated key influencing factor analysis results for the Internet of Things industry in different regions and at different development stages to improve the accuracy of the analysis.

[0118] Based on the analysis results of key influencing factors, the system generates targeted recommendations for IoT industry development. These recommendations cover multiple dimensions, including adjustments to development goals, optimization of resource allocation, breakthroughs in key areas, and improvements to coordination mechanisms. Regarding development goals, the system proposes a more coordinated approach to setting development goals based on the consistency analysis results across different levels, ensuring that the macro-development direction aligns with regional realities. Regarding resource allocation, based on the analysis of influencing factors, the system recommends prioritizing limited resources in key areas that have the greatest impact on improving consistency, such as talent development, technology research and development, and standards development. Regarding key areas, the system focuses on identified bottlenecks and proposes breakthrough solutions, such as strengthening standard development and promotion to address inconsistent technical standards. Regarding coordination mechanisms, the system analyzes differences between levels and regions and proposes specific measures to improve information communication and strengthen collaborative actions. These recommendations are not only based on data analysis but also incorporate professional knowledge and best practices in IoT industry development, making them highly practical and forward-looking. The system categorizes recommendations based on priority and implementation difficulty, creating three levels: short-term action recommendations, mid-term improvement recommendations, and long-term development recommendations, providing decision makers with a clear action guide. During actual implementation, the system continuously optimizes recommendations based on implementation feedback, forming a closed-loop improvement mechanism.

[0119] A dynamic feedback mechanism is established to continuously optimize development evaluation results. To ensure the timeliness and accuracy of evaluation methods and results, the system has established a comprehensive dynamic feedback mechanism. First, the system regularly collects new IoT industry development data and updates the text database and feature analysis results. Second, the system recalculates the PMC index based on the new data, and updates the PMC surface plot and consistency evaluation results. Third, the system tracks the implementation and effectiveness of previously proposed improvement suggestions and evaluates their effectiveness. Finally, the system continuously adjusts the evaluation algorithm parameters and recommendation generation strategy based on implementation feedback to improve the system's adaptability and accuracy. This dynamic feedback mechanism makes IoT industry development evaluation a continuous optimization process rather than a one-time static assessment. As data accumulates and algorithm optimization deepens, the system's evaluation accuracy and recommendation quality will continue to improve, providing long-term and stable decision-making support for the healthy development of the IoT industry. Furthermore, the system supports customized analysis, allowing decision-makers to set evaluation priorities and parameters based on specific needs, thereby obtaining evaluation results and improvement suggestions that are more tailored to their actual needs. This flexibility further enhances the system's practical value, enabling it to adapt to the IoT industry evaluation needs of different regions and development stages.

[0120] In one embodiment, based on the PMC index evaluation results, the process of performing a system evaluation through a preset consistency analysis model to form development and improvement suggestions may also include: Step 701: Obtain the PMC index evaluation result and establish an adaptive mind map framework.

[0121] Specifically, the Adaptive Graph of Thought (AGoT) framework is an innovative extension of the traditional Chain of Thought (CoT) and Tree of Thought (ToT) approaches. Unlike the linear CoT and branching ToT, AGoT employs a dynamic Directed Acyclic Graph (DAG) structure, which adaptively adjusts reasoning paths based on problem complexity and analysis requirements.

[0122] In IoT industry development analysis, the AGoT framework uses Graph Neural Networks (GNN) technology to construct relationships between development factors. Each development factor is represented as a node in the graph, and the dependencies between factors are represented as edges. Using the attention mechanism, the system dynamically adjusts the weights of connections between nodes, highlighting the impact of key development factors.

[0123] In this step, complete evaluation results are first obtained from the PMC index evaluation database, including the PMC index values ​​and their constituent elements for each time point and region. The Adaptive Graph of Thought (AGoT) framework is a key innovation of this invention. It breaks through the limitations of traditional linear and tree-like thinking structures and adopts a more flexible dynamic directed acyclic graph structure. In specific implementation, the system constructs an initial knowledge graph based on the PMC index evaluation results. This graph uses various elements of IoT industry development as nodes and the relationships between elements as edges to form a network structure. The system uses graph neural network technology to learn the representation of this structure, so that each node can obtain a vector representation containing rich semantic information. Unlike traditional knowledge graphs, the AGoT framework is adaptive and can dynamically adjust the graph structure and reasoning path based on the complexity of the problem and the analysis requirements. For example, when analyzing IoT technology standard consistency issues, the system automatically strengthens the weights of nodes and edges related to the standards; when analyzing industry chain collaboration issues, it adjusts to highlight the connections between upstream and downstream enterprises. This adaptive mechanism is implemented through the attention mechanism. The system calculates the importance weights of different nodes and edges based on the current analysis task to form an optimized thinking map for specific problems.

[0124] Step 702: Decompose the complex problem into interrelated sub-problems using the adaptive mind map framework.

[0125] Specifically, in this process, the AGoT framework uses recursive neural networks and the Transformer architecture to automatically decompose the problem. The system first represents the text as a high-dimensional vector and extracts semantic features using pre-trained language models such as BERT or T5.

[0126] The complex problem is then broken down into a set of sub-problems using the Question Decomposition Module (QDM). Based on a reinforcement learning strategy, QDM uses the final analysis quality as a reward signal to continuously optimize the decomposition strategy.

[0127] The decomposed sub-problems form a DAG structure, and the connection weights between nodes are dynamically adjusted through the Edge Attention Network.

[0128] Based on the established adaptive mind map framework, complex evaluation problems in IoT industry development can be decomposed into a series of interconnected but more specific sub-problems. This decomposition process is achieved by combining recursive neural networks (RNNs) and the Transformer architecture. First, the system semantically represents problematic areas identified in the PMC evaluation results (such as areas or indicators with low consistency), converting them into a high-dimensional vector space. Then, the Question Decomposition Module (QDM) breaks the complex problem into a set of sub-problems. For example, when faced with the complex issue of "lack of coordination in the IoT industry chain," the system automatically decomposes it into multiple sub-problems, such as "upstream chip supply stability," "midstream module standardization," and "downstream application scenario diversity." The QDM employs a reinforcement learning strategy, using the final analysis quality as a reward signal to continuously optimize the decomposition strategy, ensuring that the decomposed sub-problems are both relatively independent and analyzable while also covering the core elements of the original problem. During the decomposition process, the system preserves the logical connections between the sub-problems, forming a directed acyclic graph (DAG) representing the problem dependencies. This structure not only expresses the sequential dependencies between subproblems, but also avoids analytical loops caused by circular dependencies. Through this intelligent problem decomposition, the system can transform the large and complex IoT industry development evaluation problem into a manageable network of subproblems, laying the foundation for subsequent precise analysis.

[0129] Step 703: constructing a directed acyclic graph structure, prioritizing the sub-problems and analyzing the sub-problems according to the priority ranking to obtain analysis results of the sub-problems.

[0130] Specifically, the AGoT framework combines the Multi-Armed Bandit (MAB) algorithm with the Thompson sampling method to dynamically adjust the analysis priority of subproblems. For each subproblem node pi, the system maintains its expected analysis value vi and uncertainty σi.

[0131] Each time a node is selected for analysis, the system samples from the Beta distribution Beta(vi + 1, n - vi + 1), where n is the total number of attempts, and selects the node with the highest sample value for analysis. This approach strikes a balance between "exploration" (trying new analysis paths) and "exploitation" (deepening analysis of known important paths).

[0132] After decomposing the problem, the system needs to rationally arrange the analysis order of the subproblems to ensure efficiency and accuracy. The directed acyclic graph (DAG) structure constructed by the system not only represents the dependencies between subproblems but also provides a basic framework for prioritization. During the prioritization phase, the system uses a strategy combining the Multi-Armed Bandit (MAB) algorithm with the Thompson sampling method to dynamically adjust the order of analysis tasks. Specifically, for each subproblem node pi, the system maintains its expected analysis value vi (indicating the contribution of solving the subproblem to the overall analysis) and uncertainty σi (indicating the confidence level in the estimated value of the subproblem). When selecting the next subproblem to analyze, the system samples from a Beta distribution Beta(vi+1, n-vi+1), where n is the total number of attempts, and selects the subproblem with the highest sample value for analysis. This approach cleverly balances "exploration" (trying to analyze new subproblem paths) with "exploitation" (deepening analysis of known important subproblems), maximizing analytical effectiveness within limited computing resources. The system analyzes each subproblem one by one according to the determined priority order, utilizing a dedicated analysis module for each subproblem. For example, for subproblems related to technical consistency, the system invokes the standard analysis engine; for subproblems related to regional collaboration, the system invokes the regional comparison analysis engine. After each subproblem is analyzed, the system stores the results in a result pool and simultaneously updates the node status and value estimates of related dependent nodes in the directed acyclic graph to provide a reference for subsequent analysis.

[0133] Step 704: Integrate the analysis results of the sub-problems to form the development improvement suggestions.

[0134] Specifically, the AGoT framework integrates the analysis results of each sub-question to form systematic development optimization recommendations. By analyzing the dependency graph between development elements, the system can identify key nodes and weak links in the development system, thereby proposing more targeted improvement measures.

[0135] After analyzing all subproblems, these dispersed results need to be integrated into systematic development and improvement recommendations. This integration process involves more than simply concatenating the results; rather, it involves semantic understanding and reconstruction of the results through deep learning models. The system utilizes Graph Attention Networks (GAT) to process the subproblem analysis results. This network automatically adjusts the weighting and combination of different results based on the dependencies and content relevance between subproblems. The integration process is divided into three levels: first, problem-level integration. Based on the structure of a directed acyclic graph, the system integrates the results of directly related subproblems along logical dependency paths. Second, domain-level integration integrates the analysis results from different technical fields or industrial sectors across domains to identify cross-domain related issues and solutions. Finally, strategic-level integration maps the analysis results from all domains onto the strategic dimensions of IoT industry development to form comprehensive improvement recommendations. During the integration process, the system automatically detects inconsistencies and conflicts between the results of each subproblem, reconciles any conflicting content, and ensures the internal consistency of the final recommendations. Furthermore, the system supplements the theoretical basis and implementation paths for key recommendations based on the professional knowledge base of IoT industry development to enhance the actionability of the recommendations. The final development and improvement suggestions not only provide guidance on the macro-development direction, but also include detailed measures in specific areas. They take into account both short-term improvement priorities and long-term development strategies, providing all-round decision-making support for the development of the Internet of Things industry.

[0136] Step 705: Continuously update the data of the PMC index evaluation results to optimize the problem decomposition and analysis strategy.

[0137] Specifically, by establishing a dynamic feedback mechanism, the AGoT framework continuously learns from new evaluation results and continuously optimizes its problem decomposition and analysis strategies. By accumulating analytical experience, the system gradually improves the accuracy of problem diagnosis and the feasibility of optimization suggestions.

[0138] To ensure the timeliness and accuracy of the system's analysis, the present invention establishes a comprehensive dynamic update mechanism. The system regularly collects the latest IoT industry development data, updates the PMC index evaluation results, and adjusts the structure and parameters of the adaptive mind map based on the new evaluation results. During the data update phase, the system uses an incremental learning approach, recalculating only the portion of the map relevant to the new data, avoiding the computational overhead of a full map reconstruction. Simultaneously, the system records performance indicators for each problem decomposition and analysis process, such as analysis accuracy and suggestion adoption rate, and uses these indicators as reward signals for reinforcement learning to continuously optimize the problem decomposition strategy and analysis methods. As data accumulates and analytical experience grows, the system's AGoT framework will become increasingly intelligent, enabling it to more accurately capture key issues and potential opportunities in IoT industry development. Furthermore, the system incorporates an autonomous learning module that learns from feedback from industry experts on analysis results to identify improvement directions and continuously refine its knowledge base and analytical models. This continuous optimization mechanism enables the system to adapt to the rapid changes in IoT technology and industry, maintaining the forward-looking and practical nature of its analysis. In practical applications, the system supports differentiated update frequency settings, adjusting the data update cycle based on the development speed of different regions and fields to ensure efficient utilization of analytical resources.

[0139] In another embodiment, based on the PMC index evaluation results, the process of performing system evaluation using a preset consistency analysis model to form development and improvement suggestions may further include: Build a policy language agent system and establish a latent space model for policy representation.

[0140] Specifically, this module first establishes a strategic language agent system capable of understanding and generating development-related strategic language. Specifically, the system uses the PMC index evaluation results and Adaptive Map of Thought (AGoT) analysis results as input to construct a latent space for strategic representation. This latent space encompasses all possible strategies in the evaluation, optimization, and implementation processes.

[0141] In the intelligent analysis framework for IoT industry development evaluation, the policy language agent system is a key link between analysis results and specific action recommendations. The system first builds a natural language processing engine specifically designed to express and understand IoT industry strategies, transforming abstract analytical conclusions into concrete and feasible policy statements. The system leverages pretrained language models (such as BERT or T5) as its foundation. Fine-tuned on IoT literature and case data, the model is equipped to understand and generate IoT terminology and policy language. Furthermore, the system establishes a latent space model for policy representation, mapping various possible IoT development strategies into a high-dimensional continuous vector space. Unlike traditional discrete policy enumeration, the latent space model can represent an infinite number of policy variants and capture semantic similarities and combination patterns between policies. This latent space is implemented using variational autoencoder (VAE) technology. The encoder converts natural language policy statements into latent vectors, and the decoder reconstructs these latent vectors into specific policy statements. The system also incorporates a conditional control mechanism, allowing the distribution of the latent space to be specifically constrained based on specific environmental conditions (such as regional characteristics and development stage), generating strategies more suitable for specific scenarios. This latent space representation method not only greatly expands the policy space that the system can explore, but also provides a continuously differentiable optimization objective for subsequent policy optimization, making the policy generation and improvement process more flexible and efficient.

[0142] The latent space model is a hybrid generative model based on a variational autoencoder (VAE) and a generative adversarial network (GAN), specifically designed to represent and generate IoT industry development strategies. Unlike traditional discrete strategy representations, this model maps strategies into a continuous, high-dimensional latent space, capturing the semantic relationships between strategies and the potential for innovation.

[0143] The core architecture of the latent space model consists of three parts: encoder E, decoder D, and discriminator C. Encoder E maps the input policy s to a latent space vector z, i.e., z = E(s); decoder D reconstructs the latent space vector z into a policy representation s', i.e., s' = D(z); and discriminator C is responsible for distinguishing between the generated policy and the true policy, thereby improving the generated policy quality. The overall optimization objective function is: L = L recon + λ1·L KL + λ2·L adv Among them L recon is the reconstruction loss, ensuring the fidelity of the encoding-decoding process; L KL is the KL divergence loss, which makes the potential distribution close to the preset prior distribution (usually the standard normal distribution); Ladv is the adversarial loss that improves the authenticity and diversity of the generated strategy. λ1 and λ2 are weight coefficients that balance the various loss terms.

[0144] To enhance the model's understanding of IoT domain knowledge, the latent space model incorporates a domain knowledge constraint mechanism. The system constructs a knowledge graph G of IoT industry development, encompassing core concepts, relationships, and rules. During the encoding process, for an input strategy s, the system calculates its consistency score cG(s) with the knowledge graph and incorporates this score as an additional feature into the latent representation z. During the decoding process, the system adds a graph guidance layer to ensure that the generated strategy adheres to domain knowledge specifications. This mechanism significantly improves the professionalism and feasibility of the model's generated strategy.

[0145] Another innovation of the latent space model is the conditional generation mechanism. The model supports multiple conditional controls, including: (1) regional conditions cr, which adapt the generation strategy to the industrial development characteristics of different regions; (2) stage conditions cs, which generate differentiated strategies for different stages of IoT development; and (3) target conditions co, which adjust the strategy direction according to specific development goals. Conditional control is achieved by introducing conditional embeddings in the encoder and decoder. The conditional vector is concatenated with the original feature vector and then used in the calculation.

[0146] To support strategy exploration and innovation, the model incorporates a set of latent space manipulation tools, including vector interpolation (generating transition strategies between two known strategies), vector arithmetic (e.g., A - B + C, replacing a feature of strategy A with the corresponding feature of strategy C), region clustering (discovering clusters of similar strategies), and novelty search (finding innovative strategies in low-density regions). These tools enable decision makers to systematically explore the strategy space and discover innovative solutions that are difficult to access using conventional thinking.

[0147] Through this complex latent space representation, the system can handle the high-dimensional, multi-factor strategy generation problem in the development of the Internet of Things industry, and provide richer and more flexible strategy choices for decision-making.

[0148] A strategy combination is sampled from the latent space model to obtain a strategy combination, which is applied to the evaluation and optimization process.

[0149] Specifically, this module designs an iterative optimization algorithm to continuously improve the performance of the policy language agent. In each iteration, the system first samples possible policy combinations in the latent space and applies these policies to the evaluation and optimization process. By observing the effectiveness of policy execution (such as changes in the PMC index), the system learns to update the policy distribution, gradually converging to a more optimal policy space.

[0150] After having a latent space model for policy representation, the system needs to intelligently sample from this high-dimensional space to generate effective policy combinations for evaluation and optimization of IoT industry development. The system uses a particle-based policy search algorithm (PPS) for policy sampling. The algorithm maintains a set of particles {z i Each particle represents a candidate strategy point in the latent space. The system first initializes multiple particles (typically 100-500) from a prior distribution (typically a standard normal distribution). A decoder then converts each particle into a corresponding strategy representation. To improve sampling efficiency, the system employs importance sampling, adjusting the sampling distribution based on historical strategy evaluation results to focus sampling on areas with high potential returns. The system also implements a stratified sampling strategy, first sampling at the strategy category level (e.g., selecting strategy directions such as technological innovation, talent development, or standard advancement), and then refining sampling at the level of specific measures. This stratified structure ensures that the resulting strategy portfolio has both a clear strategic direction and actionable, detailed measures. The sampled strategy portfolio undergoes preliminary screening to eliminate clearly unreasonable or conflicting strategies. The portfolio is then applied to the evaluation and optimization process for IoT industry development. During application, the system simulates the effectiveness of these strategies in different scenarios, assesses their potential impact on the PMC index, and further adjusts the strategy portfolio based on the evaluation results. Through this iterative sampling-evaluation-adjustment process, the system gradually finds the optimal strategy portfolio.

[0151] A multi-level strategy feedback mechanism is established to evaluate the execution effect of the strategy combination and obtain the strategy execution evaluation results.

[0152] Specifically, this module builds a multi-layered policy feedback mechanism. The system compares policy execution results with expected goals, identifying successful and failed policy patterns. This information is used to update the distribution of the latent space, enabling the system to generate more effective policies in subsequent iterations. In particular, the system focuses on policy combinations that significantly improve the PMC index and identifies them as high-quality learning samples.

[0153] Generating a strategy portfolio is only the first step; evaluating the effectiveness of these strategies is even more crucial, providing a basis for subsequent optimization. The system establishes a multi-level strategy feedback mechanism to evaluate strategy execution across different dimensions and timescales. The system employs three primary evaluation levels: First, technical evaluation examines the strategy's impact on core IoT technology indicators, such as the rate of technological innovation, standardization, and platform compatibility; second, industrial evaluation focuses on the strategy's impact on the IoT industry structure and operational efficiency, such as industry chain synergy, enterprise innovation activity, and market application penetration; and finally, ecological evaluation assesses the strategy's impact on the health of the entire IoT ecosystem, including indicators such as diversity, stability, and sustainability. In terms of timescales, the system simultaneously conducts short-term impact assessments (within one year), medium-term impact forecasts (one to three years), and long-term impact outlooks (over three years) to comprehensively assess the strategy's immediate benefits and ongoing value. The evaluation process utilizes a combination of methods, including data-driven statistical analysis, simulation predictions based on theoretical models of IoT industry development, and qualitative judgments supported by an expert knowledge base. The system generates a detailed evaluation report for each strategy combination, including quantitative performance indicators, visual impact path analysis, and identification of key success factors. These evaluation results not only guide the adjustment and improvement of the current strategy but also serve as valuable empirical data stored in the system knowledge base, providing reference for future strategy generation.

[0154] The latent space distribution of the latent space model is updated according to the strategy execution evaluation result to generate an optimization strategy.

[0155] Specifically, the system uses a particle-based policy search algorithm (PPS) for policy sampling. The algorithm maintains a set of particles {z i}, each particle represents a policy candidate point in the latent space. The system first initializes N particles (usually N = 100-500) from the prior distribution. Then, each particle is converted into the corresponding policy representation through the decoder D(z).

[0156] Strategy evaluation uses a multi-objective evaluation function that comprehensively considers factors such as PMC index improvement, implementation cost, and time efficiency: R(z) = w1·ΔPMC(z) + w2·(1 / Cost(z)) + w3·(1 / Time(z)) Where w1, w2, w3 are weight coefficients, satisfying w1+ w2+ w3= 1.

[0157] Based on the evaluation results of policy execution, the system continuously updates and optimizes the policy generation model to generate more effective development strategies. The core of this step is to adjust the distribution parameters of the latent space model to increase the sampling probability of regions containing efficient strategies. The system employs a gradient-based policy optimization approach, using the policy evaluation score as the objective function. The system calculates the gradient direction of the latent space parameters via a backpropagation algorithm and then updates the distribution parameters along this gradient direction. To prevent premature convergence to local optima, the system introduces a temperature annealing mechanism. This mechanism maintains a high exploration temperature during the initial optimization phase and gradually decreases it as iterations increase, enhancing the utilization of regions with efficient strategies. Furthermore, the system employs ensemble learning, maintaining multiple independent latent space models, each dedicated to different types of policy generation. The outputs of these models are then combined through a weighted ensemble to enhance the diversity and robustness of policy generation. The updated latent space model generates a new generation of optimization strategies that retain the core features of the original efficient strategies while incorporating new innovative elements, forming a more comprehensive strategy system. The system regularly evaluates the performance indicators of the latent space model, such as strategy diversity, generation quality, and optimization convergence speed, and adjusts the model architecture and training parameters based on these indicators to ensure that the model is always in the best state.

[0158] Comprehensively optimize evaluation strategies, analysis strategies, and implementation strategies to improve overall performance.

[0159] Specifically, this module develops a cross-step policy optimization mechanism. It can simultaneously optimize the evaluation, analysis, and implementation strategies to improve overall performance. By sharing learned policy knowledge across different steps, the system can generate more coordinated and efficient decision recommendations.

[0160] In the complete closed-loop evaluation and optimization of IoT industry development, it's necessary not only to optimize specific development strategies but also to comprehensively optimize evaluation methods, analysis processes, and implementation mechanisms to form a systematic improvement plan. The system employs a meta-learning framework, treating evaluation, analysis, and implementation strategies as three interrelated learning tasks. This system improves overall performance through shared feature representation and collaborative optimization. Regarding evaluation strategy optimization, the system dynamically adjusts the parameter weights of the PMC index calculation model based on new data and feedback, optimizing the structural design of the indicator system and improving the accuracy and sensitivity of the evaluation. Regarding analysis strategy optimization, the system continuously refines the problem decomposition algorithm and prioritization mechanism of the AGoT framework, optimizing the dependency structure between sub-problems and increasing the efficiency and depth of analysis. Regarding implementation strategy optimization, the system improves the expressiveness and contextual understanding of the policy language agent, refines policy sampling and evaluation methods, and enhances the operability and adaptability of the generated policies. These three optimizations are not performed in isolation. The system incorporates a cross-strategy knowledge transfer mechanism to enable cross-fertilization and mutual improvement of optimization results across various links. For example, new indicator correlation patterns discovered in the evaluation strategy can guide the problem decomposition of the analysis strategy; key influencing factors identified in the analysis strategy can guide the focus of the implementation strategy; and the feedback from the implementation strategy can verify the effectiveness of the evaluation strategy. Through this cyclical improvement mechanism, the system can continuously improve itself during continued use, adapting to the new characteristics and new needs of the development of the Internet of Things industry, and providing increasingly accurate development evaluations and recommendations.

[0161] This architecture not only maintains the independence of the original steps, but also realizes the intelligent coordination of each step through the policy language agent, thereby greatly improving the overall performance of the system.

[0162] In yet another embodiment, the method further comprises: The evaluation indicators in the multi-input-output table are obtained, and a mixed linear regression model is established.

[0163] Specifically, mixed linear regression models are an effective method for handling heterogeneous and multimodal data. In IoT development data analysis, development at different levels and over different periods often exhibits distinct characteristics, making it difficult for simple linear models to accurately capture this heterogeneity. The MLR model treats data as a mixture of K different linear models, where the probability of each data point belonging to a particular model is determined by its characteristics.

[0164] The formal expression is: y = <w k , x>+ε k ~ Multinomial(π1,...,π k ) Where y is the PMC index, x is the development characteristic vector, w k is the parameter vector of the kth model, ε is the noise term, π k is the mixing weight of the k-th model.

[0165] A technique combining tensor decomposition and spectral methods is used to ensure global convergence.

[0166] Specifically, traditional hybrid linear regression methods often use the Expectation-Maximization (EM) algorithm, but are prone to falling into local optima under general data conditions. This step uses a technique that combines tensor decomposition with spectral methods to ensure global convergence under certain conditions.

[0167] Specifically, the algorithm first constructs the third-order moment tensor of the development feature:

[0168] in Represents a tensor product operation.

[0169] The parameters of each component are then estimated by tensor decomposition method:

[0170] To ensure global convergence, the algorithm introduces a sampling mechanism based on Stochastic Gradient Langevin Dynamics (SGLD):

[0171] where η t is a step sequence that satisfies Ση t =∞ and Ση t ²<∞,ξ t is standard Gaussian noise.

[0172] This mechanism can escape from local optimality in non-convex optimization problems and theoretically guarantees convergence to the global optimal solution.

[0173] The key parameters of the PMC index are estimated using a gradient-based stochastic approximation method.

[0174] Specifically, in the calculation of the PMC index, key parameters include the text strength coefficient α, the implementation period parameter β, and the consistency coefficient γ. This step uses a gradient-based stochastic approximation method to adaptively adjust these parameters through stochastic approximation technology.

[0175] This method is based on an extension of the Robbins-Monro algorithm and expresses the parameter update as:

[0176] Where θ represents the parameter vector (α, β, γ), Q is the performance metric function, and ξ t is a random disturbance.

[0177] To handle the uncertainty in parameter estimation, the system introduces an adaptive noise injection mechanism:

[0178] in is the initial step size, and a and b are hyperparameters that control the decay rate.

[0179] Use Bayesian nonparametric mixture models to handle nonlinear relationships in data.

[0180] Specifically, in order to flexibly handle the nonlinear relationships that may exist in the development data, this step introduces a Bayesian nonparametric mixed model framework. This framework uses the Dirichlet Process (DP) to automatically determine the optimal number of models K:

[0181] Where G0 is the base distribution, usually chosen to be the normal-inverse gamma distribution, and α is the concentration parameter.

[0182] Model inference adopts a method that combines Gibbs sampling and variational inference, which not only ensures inference accuracy but also improves computational efficiency.

[0183] A parameter learning algorithm is constructed to ensure the model convergence of the hybrid linear regression model, and the PMC index evaluation result containing the PMC index value is obtained.

[0184] Specifically, this step uses the Kiefer-Wolfowitz stochastic approximation method to estimate key parameters in the PMC index calculation. Compared to the fixed parameters in the original technical solution, the adaptive estimation parameters are more flexible: the text strength coefficient α is adjusted from a fixed value of {1.0, 0.8, 0.6} to a range of [0.1, 1.0] that allows for continuous variation; the implementation period parameter β is expanded from a fixed value of {2, 3, 5} to a more refined continuous estimate; and the consistency coefficient γ is also transformed from an empirically determined one to a data-driven adaptive estimate.

[0185] Alternatively, the Internet of Things industry development evaluation method provided by the present invention can be implemented by an Internet of Things industry development evaluation device. Figure 3 An embodiment of the device for evaluating the development of the Internet of Things industry of the present invention is exemplified. The device herein can be implemented by software, hardware, or a combination of both.

[0186] See Figure 3 , Figure 3This is a structural block diagram of a device for evaluating the development of the Internet of Things industry provided by an embodiment of the present invention. The device includes: The data acquisition module 801 is used to acquire data sources related to the development plan, collect and organize the data sources using text mining technology, and generate a standardized text database.

[0187] The text processing module 802 is used to perform word segmentation and stop word removal on the original text in the standardized text database, establish a text feature set by statistical word frequency distribution and feature extraction, and obtain text feature results.

[0188] The indicator construction module 803 is used to design indicators based on the text feature results and use the evaluation framework of development goals, supporting tools and supporting measures to construct a multi-input-output table.

[0189] The index calculation module 804 is used to use the evaluation indicators in the multi-input-output table to perform data processing through a preset PMC index calculation model to obtain a PMC index evaluation result.

[0190] The optimization suggestion module 805 is used to perform a system evaluation based on the PMC index evaluation results through a preset consistency analysis model to form development improvement suggestions.

[0191] The above modules can realize the functions of the corresponding steps of the method of the present invention.

[0192] In summary, the method and device for evaluating the development of the Internet of Things industry provided by the embodiments of the present invention solve the problems of strong subjectivity, low evaluation accuracy, and lack of a systematic evaluation framework in the evaluation of the development of the Internet of Things industry by constructing an evaluation framework based on the PMC index. It can provide a more comprehensive, objective, and scientific evaluation and improvement suggestions for the development of the Internet of Things industry.

[0193] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0194] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0195] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0196] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0197] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory. Memory can include non-permanent storage in a computer-readable medium, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0198] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can implement information storage in any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0199] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus comprising a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a..." does not preclude the presence of additional identical elements in the process, method, commodity, or apparatus comprising the element.

[0200] The above is a detailed introduction to the method and device for evaluating the development of the Internet of Things industry provided by the embodiment of the present invention. This article uses specific examples to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. A method for evaluating the development of the Internet of Things industry, characterized in that: include: Acquire data sources related to development planning, collect and organize the data sources using text mining technology, and generate a standardized text database; Perform word segmentation and stop word removal on the original text in the standardized text database, establish a text feature set by statistical word frequency distribution and feature extraction, and obtain text feature results; Based on the text feature results, the evaluation framework of development goals, support tools and supporting measures is used to design indicators and construct a multi-input-output table; Using the evaluation indicators in the multi-input-output table, data processing is performed through a preset PMC index calculation model to obtain a PMC index evaluation result; Based on the PMC index evaluation results, a system evaluation is conducted through a preset consistency analysis model to form development and improvement suggestions.

2. The method according to claim 1, characterized in that The data source is collected and sorted using text mining technology to generate a standardized text database, specifically including: Using keyword indexing technology to screen the data source, and generating a multi-source data set through automatic collection and processing; Performing format recognition and coding standardization processing on the multi-source data set to establish a standardized data index; Divide the data into different levels based on the normalized data index and generate a multi-level data structure through classification and sorting; Designing text identifiers using the multi-level data structure and generating a unique identifier sequence through automatic encoding; Data integration is performed based on the unique identification sequence and the multi-level data structure to generate the standardized text database.

3. The method according to claim 1, characterized in that The word segmentation and stop word removal processing of the original text in the standardized text database specifically includes: Obtaining the original text of the standardized text database, building a professional dictionary library using domain knowledge, and forming a basic word segmentation vocabulary; Perform text segmentation processing based on the basic word segmentation vocabulary to obtain an initial word segmentation result; Performing grammatical structure analysis on the initial word segmentation results and obtaining a basic phrase set through part-of-speech tagging; Using the basic phrase set to perform synonym analysis and word form unification, and obtain standard phrases through regularization processing; Stop words are identified and filtered based on the standard phrases to obtain a valid phrase set.

4. The method according to claim 1, wherein Based on the text feature results, the evaluation framework of development goals, supporting tools and supporting measures is used to design indicators and construct a multi-input-output table, including: Obtaining a feature vector of the text feature result, and using the evaluation framework to design an indicator to form an evaluation indicator framework; Design quantitative rules based on the evaluation indicator framework and establish indicator calculation standards through numerical mapping; The qualitative characteristics are numerically processed using the index calculation standard, and quantitative data are obtained through conversion rules; Designing a matrix structure based on the quantitative data to obtain a multidimensional data matrix through dimensional organization; The multidimensional data matrix is ​​used to establish an input-output mapping relationship, and a multi-input-output table is obtained through modeling.

5. The method according to claim 1, wherein The evaluation indicators in the multi-input-output table are used to process data through a preset PMC index calculation model to obtain a PMC index evaluation result, which specifically includes: The coefficient analysis is performed using the text type and the level of the issuing unit, and the text strength coefficient α value is obtained through hierarchical evaluation calculation; Analyze the implementation period data based on the text strength coefficient α value, and obtain the implementation period parameter β value through time sequence rule processing; The implementation period parameter β value is used to classify the indicators of the multi-input-output table, and the input and output variable sets are obtained through the variable identification algorithm; Performing data standardization based on the input and output variable sets, and obtaining a standardized indicator matrix through dimensional conversion; The standardized indicator matrix is ​​combined with the coefficient parameters to obtain the PMC index evaluation result containing the PMC index value through a preset PMC calculation model.

6. The method according to claim 5, characterized in that After obtaining the PMC index evaluation result including the PMC index value by using the standardized indicator matrix in combination with the coefficient parameter through a preset PMC calculation model, the method includes: Using the PMC index evaluation results to design a coordinate system, and constructing a three-dimensional display space through dimensional mapping; Performing data positioning based on the three-dimensional display space and obtaining a scattered point distribution set through spatiotemporal mapping; Performing surface fitting using the scattered point distribution set, and obtaining an initial surface using a cubic spline interpolation algorithm; Optimizing the smoothness of the initial surface and obtaining an optimized surface model by adjusting parameters; The optimized surface model is used for visualization processing, and a PMC surface graph is obtained through color mapping and node labeling.

7. The method according to claim 6, characterized in that Based on the PMC index evaluation results, a system evaluation is conducted through a preset consistency analysis model to form development and improvement suggestions, including: Based on the PMC index evaluation result, adjacent text data is extracted using the PMC surface graph, and the change rate feature is obtained by comparing the indicators; Perform hierarchical analysis based on the change rate characteristics and obtain a consistency score through matching calculation; Perform difference analysis using the consistency score and obtain difference feature vectors by index comparison; Identify influencing factors based on the difference feature vectors and obtain an influence degree matrix through correlation analysis; The influence degree matrix is ​​used to perform feature extraction, and key influencing factors are obtained through attribution analysis.

8. The method according to claim 1, characterized in that Based on the PMC index evaluation results, a system evaluation is conducted through a preset consistency analysis model to form development and improvement suggestions, including: Obtain the PMC index evaluation results and establish an adaptive mind map framework; Use the adaptive mind mapping framework to break down complex problems into interrelated sub-problems; Constructing a directed acyclic graph structure, prioritizing the subproblems and analyzing the subproblems according to the priority order, and obtaining analysis results of the subproblems; Integrate the analysis results of the sub-problems to form the development improvement suggestions; Continuously update the data of the PMC index evaluation results and optimize the problem decomposition and analysis strategy.

9. The method according to claim 1, characterized in that Based on the PMC index evaluation results, a system evaluation is conducted through a preset consistency analysis model to form development and improvement suggestions, including: Build a policy language agent system and establish a latent space model for policy representation; A strategy combination is sampled from the latent space model to obtain a strategy combination, which is applied to the evaluation and optimization process; Establish a multi-level strategy feedback mechanism to evaluate the execution effect of the strategy combination and obtain the strategy execution evaluation results; updating the latent space distribution of the latent space model according to the strategy execution evaluation result to generate an optimization strategy; Comprehensively optimize evaluation strategies, analysis strategies, and implementation strategies to improve overall performance.

10. The method according to claim 1, characterized in that include: Obtaining evaluation indicators in the multi-input-output table and establishing a mixed linear regression model; A technique combining tensor decomposition and spectral methods is used to ensure global convergence; The key parameters of the PMC index are estimated using a gradient-based stochastic approximation method; Use Bayesian nonparametric mixture models to handle nonlinear relationships in data; A parameter learning algorithm is constructed to ensure the model convergence of the hybrid linear regression model, and the PMC index evaluation result containing the PMC index value is obtained.

11. An Internet of Things industry development evaluation device, characterized in that: include: A data acquisition module is used to acquire data sources related to the development plan, collect and organize the data sources using text mining technology, and generate a standardized text database; A text processing module is used to perform word segmentation and stop word removal on the original text in the standardized text database, establish a text feature set by statistical word frequency distribution and feature extraction, and obtain text feature results; An indicator construction module is used to design indicators based on the text feature results and use an evaluation framework of development goals, supporting tools, and supporting measures to construct a multi-input-output table; An index calculation module, configured to use the evaluation indicators in the multi-input-output table to perform data processing through a preset PMC index calculation model to obtain a PMC index evaluation result; The optimization suggestion module is used to perform a system evaluation based on the PMC index evaluation results through a preset consistency analysis model to form development improvement suggestions.

Images