Text verification and quality control method for traffic transportation corpus

Abstract

The invention relates to the technical field of traffic transportation, in particular to a text verification and quality control method for a traffic transportation corpus, which comprises the following steps: S1, data acquisition: acquiring text information related to traffic transportation through multiple information sources; s2, data preprocessing: preprocessing the collected text information; s3, text classification: classifying the preprocessed text information according to different fields of transportation; s4, trust propagation based on a semantic network: analyzing the historical reliability and interaction degree of an information source to quantitatively score the credibility of each piece of information; and S5, multi-dimensional situation adaptation analysis: comprehensively considering user behaviors, environment variables and real-time traffic data, and dynamically adjusting a quality control standard of the text. According to the method, the data quality and the information accuracy can be remarkably enhanced, so that the data in the corpus can better meet the actual requirements of traffic management.

Description

Technical Field

[0001] The present invention relates to the field of transportation technology, and in particular to a text verification and quality control method for a transportation corpus. Background Art

[0002] With the rapid development of intelligent transportation systems, transportation corpora are increasingly used in traffic management, route planning, public safety, transportation efficiency and other aspects. High-quality transportation corpora provide key data support for urban traffic decision-making, intelligent navigation, real-time information release, etc., which helps to improve the efficiency of traffic management, improve road safety and optimize transportation routes. However, efficient management and accuracy of transportation corpora are crucial, because fluctuations in data quality will directly affect the effectiveness of traffic management and users' travel experience. Therefore, establishing a high-quality, dynamically adaptive text verification and quality control method has become a key technical requirement in the field of transportation.

[0003] The existing transportation corpus has many shortcomings in data quality control. First, the diversity and real-time nature of data collection are not sufficient. It usually relies on only a single or a few data sources, making it difficult to fully obtain rich traffic information. Second, the data quality control process lacks effective processing of problems such as noise, inconsistent formats, and duplicate information, resulting in low data consistency and accuracy. In addition, the existing verification mechanism usually fails to fully consider the historical reliability of the data source and the correlation between information sources, making it difficult to effectively evaluate the credibility of the information. Finally, the quality control standards are mostly fixed thresholds, lacking the ability to dynamically adjust according to real-time traffic conditions, environmental variables, and user needs, resulting in poor responsiveness and adaptability of the system in different scenarios.

[0004] The purpose of the present invention is to provide a text verification and quality control method for a transportation corpus, dynamically adjust the quality control standard of the text, thereby effectively improving the practicality, reliability and adaptability of the data, and providing strong support for traffic management and user decision-making. Summary of the invention

[0005] The present invention provides a text verification and quality control method for a transportation corpus.

[0006] The text verification and quality control method of the transportation corpus includes the following steps: S1, data collection: collect text information related to transportation through various information sources (such as traffic monitoring platforms, user-submitted data, social media, etc.); S2, data preprocessing: preprocessing the collected text information, including removing noise, standardizing the format, and eliminating duplicate content; S3, text classification: classify the preprocessed text information according to different areas of transportation (traffic safety, route planning, transportation efficiency); S4, trust propagation based on semantic web: After text classification, the relationship map between information sources is constructed to analyze the historical reliability and interaction level of information sources to quantitatively score the credibility of each piece of information, including: S41, information source relationship map construction: analyze the classified text information and construct the relationship map between information sources; S42, historical reliability analysis of classified information: for the classified text information, the reliability of each information source is evaluated using the historical text information, and an initial trust value is assigned to the node of each information source; S43, Trust Propagation and Score Calculation: Apply the trust propagation algorithm based on the semantic web to iteratively update the trust value of each node; S5, Multi-dimensional Context Adaptation Analysis: Using the multi-dimensional context adaptation analysis algorithm, the quality control standards of the text are dynamically adjusted by comprehensively considering user behavior, environmental variables (weather, traffic flow) and real-time traffic data.

[0007] Optionally, the data collection in S1 includes: S11, traffic status information acquisition: obtain traffic flow, accident reports and road condition update data in real time through the traffic monitoring platform to obtain the latest traffic status information; S12, user submitted information collection: receiving feedback, complaints and suggestions submitted by users through mobile applications and online platforms, and collecting user submitted information related to transportation; S13, social media information crawling: crawl public posts, comments and real-time dynamics related to transportation from social media platforms to obtain social media information related to transportation.

[0008] Optionally, the data preprocessing in S2 includes: S21, noise removal: Apply text cleaning algorithm to remove irrelevant characters and noise data; S22, format standardization: unify the format of the text, including case conversion, punctuation standardization and word standardization; S23, duplicate content elimination: Use hash algorithm to deduplicate text and identify and remove duplicate content by calculating the hash value of the text.

[0009] Optionally, the text classification in S3 includes: S31, feature extraction: extract the feature vector of the text using the TF-IDF method; S32, text information classification: Use the support vector machine model to classify the extracted text feature vectors and categorize the text information according to different areas of transportation (such as traffic safety, route planning, and transportation efficiency).

[0010] Optionally, the information source relationship graph construction in S41 includes: S411, information source node identification: by analyzing the classified text information, identifying the text information from different sources, and treating each information source (such as traffic monitoring system, user submission platform, social media, etc.) as a node in the graph ; S412, relationship edge weight calculation: Determine the edge weights between nodes by calculating the relevance between information sources through similar text feature vectors ; S413, Relationship graph construction: Use node and edge weights to construct a relationship graph between information sources ,in represents the information source node set, represents the set of connecting edges between information sources, Represents the weight of the edge.

[0011] Optionally, the historical reliability analysis of the classification information in S42 includes: S421, historical accuracy calculation: by analyzing the historical text information of the information source under the category, the accuracy of the information released by it is calculated; S422, information consistency calculation: based on the consistency of historical information of the information source under the same category, the cosine similarity is used to calculate the similarity of information released by the same information source in different periods, and its stability is evaluated; S423, initial trust value calculation: assign an initial trust value to each information source based on historical accuracy and information consistency.

[0012] Optionally, the trust propagation and score calculation in S43 includes: S431, trust propagation calculation: In each round of iteration, the trust value is updated according to the connection relationship and association strength of each information source node through the trust propagation algorithm. ; S432, iterative termination condition: repeat the trust propagation calculation until the trust values ​​of all nodes converge, that is, satisfy ,in is the preset convergence threshold.

[0013] Optionally, the multidimensional situational adaptation analysis in S5 includes: S51, context variable collection and analysis: Collect multidimensional context variables related to text quality, including user behavior, environmental variables, and real-time traffic data, and analyze the multidimensional context variables to evaluate the impact of different variables on text quality requirements; S52, Dynamic adjustment of quality control standards: Dynamically adjust the quality control standards of the text based on the analysis results of multidimensional situational variables.

[0014] Optionally, the situational variable collection and analysis in S51 includes: S511, Situational variable collection: Collect multi-dimensional situational variables, including user behavior data, through sensor monitoring and external data interfaces , Environment variables and real-time traffic data ; S512, variable weight analysis: Use multi-dimensional situational adaptation analysis algorithm to perform weighted analysis on situational variables, and calculate the comprehensive quality demand impact value of situational variables. , evaluate the impact of each variable on text quality requirements.

[0015] Optionally, the dynamic adjustment of the quality control standard in S52 includes: S521, determine the quality requirement level: the comprehensive quality requirement impact value The upper limit of the preset quality threshold , quality threshold lower limit For comparison, When When , indicating a high demand level; S522, Adjust quality control standards: Dynamically adjust quality control standards according to different quality requirement levels, including: Low demand level adjustment: adopt basic quality control standards; Medium demand level adjustment: improve the accuracy and update frequency of text information; High-demand level adjustment: Set the quality control standard to the highest level, requiring real-time and accuracy of text, increasing the frequency of information updates, and strengthening data verification.

[0016] Beneficial effects of the present invention: The present invention realizes efficient organization and credibility assessment of information through systematic steps, including data collection, preprocessing, classification and trust propagation. Data collection covers a variety of information sources, ensuring the richness and diversity of information. Data preprocessing further improves the purity, consistency and uniqueness of information, and can significantly enhance data quality and information accuracy, so that the data in the corpus can better serve the actual needs of traffic management.

[0017] The present invention accurately quantifies the credibility of each information source by constructing an information source relationship map and utilizing historical reliability analysis of classified information. The iterative calculation and dynamic adjustment of trust propagation ensure that the correlation strength between information sources can be reasonably reflected in the trust value, which not only reveals the potential reliability of the data, but also makes the final scoring results more comprehensive and accurate, thereby enhancing the credibility of corpus data in the field of transportation.

[0018] The present invention dynamically adjusts the quality control standards of the text by collecting and weightedly analyzing user behavior, environmental variables and real-time traffic data. According to different situational requirements, the system can adjust the accuracy, update frequency and data verification requirements of the text in real time to ensure that the text information remains highly practical and reliable in a changing environment. This not only improves the system's responsiveness and resource utilization efficiency, but also enables the text content to accurately meet the needs of users in complex traffic situations, providing efficient, stable and adaptable support for traffic management and user decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A schematic diagram of a control method flow chart of an embodiment of the present invention; Figure 2 Schematic diagram of trust propagation based on the semantic web according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0022] It should be noted that the references to "one embodiment", "embodiment", "exemplary embodiments", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0023] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0024] like Figure 1-Figure 2 As shown in FIG. 1 , the text verification and quality control method of the transportation corpus includes the following steps: S1, data collection: collect text information related to transportation through various information sources (such as traffic monitoring platforms, user-submitted data, social media, etc.); S2, data preprocessing: preprocessing the collected text information, including removing noise, standardizing the format, and eliminating duplicate content; S3, text classification: classify the preprocessed text information according to different areas of transportation (traffic safety, route planning, transportation efficiency); S4, trust propagation based on semantic web: After text classification, the relationship map between information sources is constructed to analyze the historical reliability and interaction level of information sources to quantitatively score the credibility of each piece of information, including: S41, information source relationship map construction: analyze the classified text information and construct the relationship map between information sources; S42, historical reliability analysis of classified information: for the classified text information, the reliability of each information source is evaluated using the historical text information, and an initial trust value is assigned to the node of each information source; S43, Trust Propagation and Score Calculation: Apply the trust propagation algorithm based on the semantic web to iteratively update the trust value of each node; S5, Multi-dimensional Context Adaptation Analysis: Using the multi-dimensional context adaptation analysis algorithm, the quality control standards of the text are dynamically adjusted by comprehensively considering user behavior, environmental variables (weather, traffic flow) and real-time traffic data; Through the above content, the high quality and credibility of text information are ensured. Combined with multi-dimensional situational adaptation analysis, the quality control standards can be dynamically adjusted to reflect user needs and environmental changes in real time, thereby effectively improving the practicality and reliability of information. It provides strong support for data management in the transportation field and helps to improve decision-making efficiency and service quality.

[0025] Data collection in S1 includes: S11, traffic status information acquisition: obtain traffic flow, accident reports and road condition update data in real time through the traffic monitoring platform to obtain the latest traffic status information; S12, user submitted information collection: receiving feedback, complaints and suggestions submitted by users through mobile applications and online platforms, and collecting user submitted information related to transportation; S13, social media information capture: capture public posts, comments and real-time dynamics related to transportation from social media platforms to obtain social media information related to transportation; Through the above content, traffic status information can be obtained in real time, user feedback and needs can be reflected in a timely manner, and trends and public sentiment on social media can be captured, thereby improving the richness and diversity of data. It not only enhances the reliability of the data, but also provides a more comprehensive perspective for subsequent analysis and decision-making, supporting the effectiveness and adaptability of transportation management.

[0026] Data preprocessing in S2 includes: S21, noise removal: Apply text cleaning algorithm to remove irrelevant characters and noise data; S22, format standardization: unify the format of the text, including case conversion, punctuation standardization and word standardization; S23, duplicate content elimination: Use hash algorithm to deduplicate texts, and identify and remove duplicate content by calculating the hash value of the text; Through the above content, the quality and consistency of collected text information are improved, noise removal ensures the purity of data, format standardization improves the consistency of information, and elimination of duplicate content ensures the uniqueness of the text, thereby avoiding the interference of redundant information on subsequent analysis. It not only provides a reliable foundation for subsequent text classification and quality control, but also enhances the accuracy and effectiveness of data analysis, thereby improving the decision-making support capabilities of transportation management.

[0027] Text classification in S3 includes: S31, feature extraction: The TF-IDF method is used to extract the feature vector of the text, expressed as: ; in, Expressive words In the documentation The frequency of occurrence in Expressive words The inverse document frequency of S32, text information classification: Use the support vector machine model to classify the extracted text feature vectors, and classify the text information according to different fields of transportation (such as traffic safety, route planning, and transportation efficiency), expressed as: ; in, is the classification result, is the weight vector of the support vector machine, is the feature vector to be classified, is the bias term, It is a symbolic function used to determine the final classification result; Through the above content, the text information related to transportation is accurately divided into fields such as traffic safety, route planning and transportation efficiency. This not only improves the accuracy and consistency of text classification, but also ensures that the information can meet the needs of different fields in a targeted manner. By introducing specific classification standards, this method enhances the practicality of the corpus in traffic management and decision support, so that the data can more effectively serve specific transportation application scenarios.

[0028] The information source relationship graph construction in S41 includes: S411, information source node identification: by analyzing the classified text information, identifying the text information from different sources, and treating each information source (such as traffic monitoring system, user submission platform, social media, etc.) as a node in the graph ; S412, relationship edge weight calculation: Determine the edge weights between nodes by calculating the relevance between information sources through similar text feature vectors , expressed as: ; ; in, Indicates the source of information and The similarity of the text content provided, and Information sources and The text feature vector of S413, Relationship graph construction: Use node and edge weights to construct a relationship graph between information sources ,in represents the information source node set, represents the set of connecting edges between information sources, represents the weight of the edge; Through the above content, the potential connections and data sharing levels between information sources can be effectively revealed, providing a solid foundation for subsequent trust propagation and credibility assessment. Using this map, the reliability of information sources can be more accurately assessed, thereby improving the overall quality and credibility of the data in the corpus.

[0029] The historical reliability analysis of the classified information in S42 includes: S421, historical accuracy calculation: by analyzing the historical text information of the information source under the category, the accuracy of its published information is calculated, which is expressed as: ; in, Indicates the source of information The number of correct information published historically under this category, Indicates the total amount of historical information released by the information source; S422, information consistency calculation: Based on the historical information consistency of the information source under the same category, the cosine similarity is used to calculate the similarity of information released by the same information source in different periods, and its stability is evaluated, which is expressed as: ; in, Indicates the source of information In the The text feature vector of the time period, Indicates the total number of time periods, is the cosine similarity function; S423, initial trust value calculation: assign an initial trust value to each information source based on historical accuracy and information consistency, expressed as: ; in, Represented as the initial trust value, and is the weight parameter, satisfying ; Through the above content, using the historical performance of information sources, not only the reliability and stability of information can be effectively evaluated, but also sources with higher credibility can be identified, ensuring the objectivity and accuracy of information source evaluation, thereby improving the credibility of the overall corpus data and the reliability of the analysis results.

[0030] Trust propagation and score calculation in S43 include: S431, trust propagation calculation: In each round of iteration, the trust value is updated according to the connection relationship and association strength of each information source node through the trust propagation algorithm. , expressed as: ; in, Indicates Node after round iteration The trust value, is the initial trust value, is the attenuation coefficient of trust propagation, which is used to control the weight of external influence. , Representation and Node The set of directly connected nodes, Representation Node and The edge weights between them (i.e., the strength of the relationship between information sources); S432, iterative termination condition: repeat the trust propagation calculation until the trust values ​​of all nodes converge, that is, satisfy ,in is the preset convergence threshold; Convergence Threshold It is dynamically set according to the number of information sources and the range of trust values, expressed as: ; in, It is a tuning parameter, usually between 0.01 and 0.1, used to control the convergence accuracy. is the total number of information source nodes, Indicates After iterations, nodes The trust value, Indicates After iterations, nodes Trust value; Through the above content, the trust value of the information source can be effectively transmitted and adjusted to ensure that the final score fully reflects the reliability of the information source. At the same time, the setting of the dynamic convergence threshold ensures the convergence speed and calculation accuracy of the algorithm, making the scoring results more accurate and efficient, thus providing a stable and reliable basic support for the text verification and quality control of the transportation corpus.

[0031] The multidimensional situational adaptation analysis in S5 includes: S51, context variable collection and analysis: Collect multidimensional context variables related to text quality, including user behavior, environmental variables, and real-time traffic data, and analyze the multidimensional context variables to evaluate the impact of different variables on text quality requirements; S52, Dynamic adjustment of quality control standards: Dynamically adjust the quality control standards of the text based on the analysis results of multidimensional situational variables; Through the above content, the accuracy and relevance of text information can be improved in real time according to different situations (such as weather changes, traffic congestion, etc.), which not only improves the practicality and reliability of the data, but also ensures that the text content can accurately meet the user's information needs in a changing traffic environment, thereby providing more accurate support for traffic management and user decision-making.

[0032] The collection and analysis of contextual variables in S51 include: S511, Situational variable collection: Collect multi-dimensional situational variables, including user behavior data, through sensor monitoring and external data interfaces , Environment variables and real-time traffic data , including: User behavior data: user access frequency , Dwell time , Feedback Rating ; Environmental variables: weather conditions (sunny, rainy, snowy, etc.) and traffic flow ; Real-time traffic data: such as current traffic speed and congestion index ; S512, variable weight analysis: Use multi-dimensional situational adaptation analysis algorithm to perform weighted analysis on situational variables, and calculate the comprehensive quality demand impact value of situational variables. , evaluate the impact of each variable on text quality requirements, expressed as: ; in, , , is the weight parameter, satisfying ; Through the above content, precise adaptation to different situational needs is achieved, high-demand situations can be dynamically identified, and text information is ensured to remain highly practical and accurate in various traffic environments, providing users with more practical decision support and effectively improving the responsiveness and reliability of data management.

[0033] Dynamic adjustment of quality control standards in S52 includes: S521, determine the quality requirement level: the comprehensive quality requirement impact value The upper limit of the preset quality threshold , quality threshold lower limit For comparison, When When , indicating a high demand level; Upper quality threshold , quality threshold lower limit Calculate the comprehensive quality demand impact value of situational variables based on historical data The mean and standard deviation are set as: ; ; in, yes The average value of yes The standard deviation of S522, Adjust quality control standards: Dynamically adjust quality control standards according to different quality requirement levels, including: Low demand level adjustment: adopt basic quality control standards, focusing on the basic accuracy and completeness of information; Adjustment of the medium demand level: Improve the accuracy and update frequency of text information to ensure that the text can reflect the current traffic conditions in a timely manner; High-demand level adjustment: Set the quality control standard to the highest level, requiring the timeliness and accuracy of the text, increasing the frequency of information updates, and strengthening data verification to ensure the reliability of the data; The basic quality control standards include: Accuracy threshold: The basic accuracy requirement is set at 95%, which means that the system only needs to verify whether the main information of the text content is accurate, without strict in-depth verification; Completeness requirements: The completeness rate of information fields is set to 100%, that is, all key fields must be filled in (such as location, time, event type, etc.) to ensure that the basic information is complete.

[0034] Update frequency: Update once an hour (or less frequently), and adjust dynamically according to the situation to avoid unnecessary resource consumption; Through the above content, an optimal balance between system resources and information quality is achieved. At low demand levels, the basic standards ensure the basic accuracy and completeness of information, while at high demand levels, the real-time and accuracy of information are strictly controlled. This not only improves the system's responsiveness and efficiency, but also ensures that text information remains practical and reliable in different situations, providing stable and adaptable support for data management and decision support of the transportation corpus.

[0035] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0036] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A text verification and quality control method for a transportation corpus, characterized in that: The following steps are involved: S1, data collection: collect text information related to transportation through various information sources; S2, data preprocessing: preprocessing the collected text information, including removing noise, standardizing the format, and eliminating duplicate content; S3, text classification: classify the preprocessed text information according to different fields of transportation; S4, trust propagation based on semantic web: After text classification, the relationship map between information sources is constructed to analyze the historical reliability and interaction level of information sources to quantitatively score the credibility of each piece of information, including: S41, information source relationship map construction: analyze the classified text information and construct the relationship map between information sources; S42, historical reliability analysis of classified information: for the classified text information, the reliability of each information source is evaluated using the historical text information, and an initial trust value is assigned to the node of each information source; S43, Trust Propagation and Score Calculation: Apply the trust propagation algorithm based on the semantic web to iteratively update the trust value of each node; S5, Multi-dimensional Context Adaptation Analysis: Using the multi-dimensional context adaptation analysis algorithm, the quality control standards of the text are dynamically adjusted by comprehensively considering user behavior, environmental variables and real-time traffic data.

2. The text verification and quality control method of the transportation corpus according to claim 1 is characterized in that: The data collection in S1 includes: S11, traffic status information acquisition: obtain traffic flow, accident reports and road condition update data in real time through the traffic monitoring platform to obtain the latest traffic status information; S12, user submitted information collection: receiving feedback, complaints and suggestions submitted by users through mobile applications and online platforms, and collecting user submitted information related to transportation; S13, social media information crawling: crawl public posts, comments and real-time dynamics related to transportation from social media platforms to obtain social media information related to transportation.

3. The text verification and quality control method of the transportation corpus according to claim 1 is characterized in that: The data preprocessing in S2 includes: S21, noise removal: Apply text cleaning algorithm to remove irrelevant characters and noise data; S22, format standardization: unify the format of the text, including case conversion, punctuation standardization and word standardization; S23, duplicate content elimination: Use hash algorithm to deduplicate text and identify and remove duplicate content by calculating the hash value of the text.

4. The text verification and quality control method of the transportation corpus according to claim 1 is characterized in that: The text classification in S3 includes: S31, feature extraction: extract the feature vector of the text using the TF-IDF method; S32, text information classification: Use the support vector machine model to classify the extracted text feature vectors and categorize the text information according to different fields of transportation.

5. The text verification and quality control method of the transportation corpus according to claim 1 is characterized in that: The information source relationship graph construction in S41 includes: S411, information source node identification: by analyzing the classified text information, identifying text information from different sources, and treating each information source as a node in the graph ; S412, relationship edge weight calculation: Determine the edge weights between nodes by calculating the relevance between information sources through similar text feature vectors ; S413, Relationship Graph Construction: Use node and edge weights to construct a relationship graph between information sources ,in represents the set of information source nodes, represents the set of connecting edges between information sources, Represents the weight of the edge.

6. The text verification and quality control method of the transportation corpus according to claim 5 is characterized in that: The historical reliability analysis of the classification information in S42 includes: S421, historical accuracy calculation: by analyzing the historical text information of the information source under the category, the accuracy of the information released by it is calculated; S422, information consistency calculation: based on the consistency of historical information of the information source under the same category, the cosine similarity is used to calculate the similarity of information released by the same information source in different periods, and its stability is evaluated; S423, initial trust value calculation: assign an initial trust value to each information source based on historical accuracy and information consistency.

7. The text verification and quality control method of the transportation corpus according to claim 6 is characterized in that: The trust propagation and score calculation in S43 include: S431, trust propagation calculation: In each round of iteration, the trust value is updated according to the connection relationship and association strength of each information source node through the trust propagation algorithm. ; S432, iterative termination condition: repeat the trust propagation calculation until the trust values ​​of all nodes converge, that is, satisfy ,in is the preset convergence threshold.

8. The text verification and quality control method of the transportation corpus according to claim 1 is characterized in that: The multi-dimensional situational adaptation analysis in S5 includes: S51, context variable collection and analysis: Collect multidimensional context variables related to text quality, including user behavior, environmental variables, and real-time traffic data, and analyze the multidimensional context variables to evaluate the impact of different variables on text quality requirements; S52, Dynamic adjustment of quality control standards: Dynamically adjust the quality control standards of the text based on the analysis results of multidimensional situational variables.

9. The text verification and quality control method of the transportation corpus according to claim 8 is characterized in that: The situation variable collection and analysis in S51 includes: S511, Situational variable collection: Collect multi-dimensional situational variables, including user behavior data, through sensor monitoring and external data interfaces , Environment variables and real-time traffic data ; S512, variable weight analysis: Use multi-dimensional situational adaptation analysis algorithm to perform weighted analysis on situational variables, and calculate the comprehensive quality demand impact value of situational variables. , evaluate the impact of each variable on text quality requirements.

10. The text verification and quality control method of the transportation corpus according to claim 9 is characterized in that: The dynamic adjustment of the quality control standard in S52 includes: S521, determine the quality requirement level: the comprehensive quality requirement impact value The upper limit of the preset quality threshold , quality threshold lower limit For comparison, When When , indicating a high demand level; S522, Adjust quality control standards: Dynamically adjust quality control standards according to different quality requirement levels, including: Low demand level adjustment: adopt basic quality control standards; Medium demand level adjustment: improve the accuracy and update frequency of text information; High-demand level adjustment: Set the quality control standard to the highest level, requiring real-time and accuracy of text, increasing the frequency of information updates, and strengthening data verification.

Images