A new source evaluation method for internet technology information

Abstract

This invention discloses a new source evaluation method for internet technology information, belonging to the field of new source evaluation technology. To address the problems in existing technologies, such as users struggling to quickly find valuable information from a large volume of technology news, significant differences in information quality between different sources, and the lack of an effective source evaluation system, which prevents users from filtering information based on source reputation, this new source evaluation method for internet technology information effectively improves information acquisition and processing efficiency, ensures content quality and credibility, enhances readability and usability through sentiment analysis, topic classification, and relevance calculation, supports users in making informed decisions, promotes a virtuous cycle in the information ecosystem through multi-dimensional considerations in the evaluation system, and dynamically optimizes source evaluation through a user feedback mechanism, enabling personalized content recommendations, meeting user needs, and improving satisfaction.

Description

Technical Field

[0001] This invention relates to the field of new information source evaluation technology, and in particular to a new information source evaluation method for Internet technology information. Background Technology

[0002] With the rapid development of internet technology, a massive amount of scientific and technological information floods the internet, leading to information overload for users. Traditional information retrieval and filtering methods are no longer sufficient to meet users' demands for high-quality, highly relevant scientific and technological information. For example, Chinese patent CN115900712B discloses a source credibility evaluation combined positioning method, which includes: acquiring scene features of the environment in which the target object is located; acquiring signal features of multiple sources for the target object; constructing a target feature vector based on the scene features and the signal features, the target feature vector reflecting the scene features and the signal features; inputting the target feature vector into a target machine learning model to obtain the credibility of each source among the multiple sources; determining a target source combination for positioning based on the credibility of each source among the multiple sources, the target source combination including at least one source among the multiple sources; determining a combined positioning model based on the target source combination; and using the combined positioning model to locate the target object.

[0003] The aforementioned patents primarily address the issues of inconsistent source quality and a lack of source evaluation; however, the following problems still exist:

[0004] In existing technologies, users find it difficult to quickly find valuable information from a large amount of scientific and technological information, and the quality of information provided by different sources varies greatly. Furthermore, there is a lack of an effective source evaluation system, making it impossible for users to filter information based on the reputation of the source. Summary of the Invention

[0005] The purpose of this invention is to provide a new information source evaluation method for Internet technology information. By constructing a new information source screening mechanism, it automatically identifies and filters potential high-quality information sources, establishes an information source evaluation system to score and rank information sources, and allows users to select information sources based on the credit rating of the information sources, thereby solving the problems mentioned in the background art.

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

[0007] A new method for evaluating information sources in internet technology news includes the following steps:

[0008] Step 1: Acquisition of new information sources: Automatically capture relevant new information source content based on the constructed new information source filtering mechanism, preprocess the captured new information source content, establish a new information source database, and store the captured new information source content;

[0009] Step Two: New Source Content Analysis: Extract key information from the new source information and analyze the new source information content using natural language processing technology;

[0010] Step 3: Evaluation of New Information Sources: Construct an evaluation system based on the analysis results of the new information sources, score the information sources according to the evaluation system, and rank the information sources according to their scores.

[0011] Step 4: Output Results: Based on the scoring results, the information sources are divided into different credit levels, and then the information sources are arranged from high to low according to the credit level.

[0012] Furthermore, the new information source filtering mechanism constructed in step one specifically includes:

[0013] Identify the target area: Define the specific area of ​​the new information source and select relevant scientific and technological information keywords based on the characteristics of the new information source area;

[0014] Data crawling: Based on the selected science and technology information keywords, use web crawling technology to send HTTP requests to Internet platforms to obtain the HTML source code of the web pages, parse the web page content, and extract the links corresponding to each Internet platform as seed links of potential information sources.

[0015] Establish screening rules: Set the update frequency of information sources according to the timeliness requirements of new information source fields, comprehensively evaluate the content quality of information sources based on the quality influencing factors of information sources, and screen information sources that match the needs and characteristics of the target audience.

[0016] Furthermore, the data crawling also includes: setting a scheduled task, running the crawler program periodically to access the target website based on the set scheduled task, the crawler program identifying and recording the timestamp of the last crawling operation, comparing the currently obtained content with the information recorded during the last crawling, and only crawling and collecting the content added and updated since the last crawling.

[0017] Furthermore, the preprocessing of the captured information in step one specifically includes:

[0018] Obtain and identify HTML tags in the webpage, obtain the information attributes of the HTML tags, divide the content data referenced by the HTML tags according to the information attributes, and obtain a content data distribution map;

[0019] At the same time, it identifies and removes advertising content from web pages based on the advertising recognition database;

[0020] Based on the content data distribution map, the main content area in the page is identified, the metadata information of the main content is extracted, and the extracted information content and its metadata are processed into data structure.

[0021] Furthermore, the new information source database includes a classification storage and identification model for information content tables, metadata tables, information source evaluation tables, and user feedback tables, and the acquired data is classified and stored based on the classification storage and identification model.

[0022] Furthermore, the analysis of the new information source content in step two specifically includes:

[0023] Keyword extraction: Based on the extracted metadata information, identify the title of the main content, summarize the main content, and extract the target keywords from the main content;

[0024] Natural Language Processing Analysis: Based on sentiment analysis technology, the sentiment tendency of new information sources is determined. Based on target keywords, the new information sources are classified into themes. At the same time, information clusters with the same theme are generated by integrating the content similarity between new information sources.

[0025] Content quality assessment: Based on the extracted metadata information, the author, publishing organization information and publication date are identified to detect whether the new source information is original, and it is compared with authoritative information to determine the accuracy of the new source information.

[0026] Source release pattern analysis: Analyze the release time of information sources to see if there are peak release periods, calculate the release frequency of information sources, and evaluate their activity level based on the calculation results;

[0027] Trend Analysis: Identify trending topics by analyzing the repetition intensity of titles and keywords in the main content. Based on historical data, predict potential new news source trends and hot topics in the future. Also, analyze the breadth and depth of news source reports, including the diversity of topics and geographical coverage.

[0028] Furthermore, the extraction of target keywords from the main content specifically includes:

[0029] From the summary of the main content, continuous word combinations are extracted based on contextual relationships. The extracted word combinations are then concatenated according to grammatical rules and semantics to obtain concatenated sentences.

[0030] The concatenated sentences are tagged with parts of speech, and the sentences are cleaned based on part-of-speech statistical features and domain-specific terminology to obtain a keyword set;

[0031] Each word in the keyword set is converted into a corresponding word vector. The distance between each word vector and its corresponding standard word vector is calculated. The keyword corresponding to the word vector with the smallest distance is selected as the target keyword.

[0032] Furthermore, generating information clusters on the same topic also includes:

[0033] Analyze the characteristics of each information cluster, and determine one or more ranking factors associated with the topic of the information cluster based on the information cluster characteristics;

[0034] Conduct content similarity analysis between information clusters to identify common keywords and themes in different information clusters, and construct an association network of information clusters based on the similarity analysis results;

[0035] A weight is assigned to each ranking factor, and a comprehensive score is calculated for each information cluster based on the ranking factor and weight. The corresponding information clusters are ranked according to the comprehensive score, and the degree of correlation between each information cluster and other topics is marked next to each information cluster.

[0036] Furthermore, the evaluation of new information sources in step three also includes:

[0037] Establish a user feedback mechanism, collect user feedback data on the information source evaluation system at preset time intervals, and classify the collected feedback data;

[0038] Identify positive and negative sentiments in user feedback data and adjust evaluation criteria and weights based on user feedback.

[0039] Furthermore, the output of the result in step four specifically includes:

[0040] Based on the scoring results, a clear correspondence between the scoring score range and the credit rating is established, and the information sources are classified into the corresponding credit ratings based on the corresponding keywords.

[0041] Information sources are sorted according to credit rating and score. For information sources with the same credit rating, a secondary sort is performed based on the specific score.

[0042] Compared with the prior art, the beneficial effects of the present invention are:

[0043] This invention proposes a new method for evaluating internet technology information sources, effectively improving the efficiency of information acquisition and processing while ensuring content quality and credibility. Through sentiment analysis, topic classification, and relevance calculation, it enhances the readability and usability of content, supporting users in making informed decisions. The multi-dimensional considerations of the evaluation system promote a virtuous cycle in the information ecosystem, while the user feedback mechanism dynamically optimizes the source evaluation. Ultimately, it achieves personalized content recommendations, meeting user needs and improving satisfaction. This method not only improves information transparency but also drives overall progress in the field of technology information. Attached Figure Description

[0044] Figure 1 This is a flowchart illustrating the steps of the new information source evaluation method for internet technology information of the present invention. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] To address the technical challenges in existing technologies, such as users struggling to quickly find valuable information from a vast amount of scientific and technological information, significant differences in information quality between different sources, and the lack of an effective source evaluation system that prevents users from filtering information based on source reputation, please refer to [link to relevant documentation]. Figure 1 This embodiment provides the following technical solution:

[0047] A new method for evaluating information sources in internet technology news includes the following steps:

[0048] Step 1: Acquisition of New Information Sources: Determine the preset keywords and target areas for internet technology information. Using web crawling technology, crawl information from internet platforms such as news, blogs, forums, and social media within the target area to obtain seed links for potential information sources, including news reports, professional new information sources, and industry reports. Set information source filtering rules, such as update frequency, content quality, and target audience. Automatically crawl relevant new information source content based on the constructed new information source filtering mechanism. Preprocess the crawled new information source content, such as deduplication, ad filtering, and text extraction. Establish a new information source database and store the crawled new information source content.

[0049] The new information source database includes a classification storage and identification model for information content tables, metadata tables, information source evaluation tables, and user feedback tables. The acquired data is classified and stored based on the classification storage and identification model.

[0050] In this embodiment, the captured new information source information is preprocessed, specifically including:

[0051] Obtain and identify HTML tags in the webpage, obtain the information attributes of the HTML tags, divide the content data referenced by the HTML tags according to the information attributes, and obtain a content data distribution map;

[0052] HTML tags define the structure and content of a webpage, including: webpage title, heading elements, paragraph elements, link elements, images and text content elements, section elements, etc.

[0053] At the same time, it identifies and removes advertising content on web pages based on the advertising recognition database, including links, images, text descriptions, etc.

[0054] Based on the content data distribution map, the main content area in the page is identified, and metadata information such as title, author, publication date, and source of the main content is extracted. The extracted information content and its metadata are then processed into data structure, such as by converting them into JSON or XML format.

[0055] Step Two: New Source Content Analysis: Extract key information from the new source information, such as title, abstract, keywords, and author. Analyze the new source information using natural language processing technology, including sentiment analysis, topic classification, and relevance calculation. At the same time, analyze the source's publishing patterns, such as publication time, frequency, and content coverage.

[0056] Step 3: Evaluation of New Information Sources: Based on the analysis results of new information source content, an evaluation system is constructed, including multiple dimensions such as content quality, timeliness, authority, and audience feedback. The information sources are scored according to the evaluation system and ranked according to their scores.

[0057] Establish a user feedback mechanism to collect user feedback data on the information source evaluation system at preset time intervals, and classify the collected feedback data, such as by information source, evaluation dimension, user satisfaction, etc., identify positive and negative emotions in user feedback data, and adjust the evaluation criteria and weights based on user feedback.

[0058] Step 4: Results Output: Based on the rating results, information sources are divided into different credit levels and ranked from highest to lowest credit level, with higher-credit-level sources listed first for easier user selection. The credit evaluation results are displayed using charts, heatmaps, etc., allowing users to intuitively understand the quality and ranking of each information source and quickly select high-quality sources. Specifically, this includes:

[0059] Based on the scoring results, a clear correspondence is established between the scoring score range and the credit rating. The information sources are categorized into the corresponding credit rating according to the corresponding keywords. For example: 9-10 points: excellent; 7-8 points: good; 5-6 points: average; 3-4 points: poor; 1-2 points: very poor. The information sources are sorted according to the credit rating and score. For information sources of the same credit rating, a secondary sort is performed based on the specific scoring score.

[0060] In this embodiment, by pre-setting keywords and target domains and combining them with web crawling technology, relevant internet technology information can be efficiently discovered and crawled, saving time spent on manual searching and filtering. Setting source filtering rules ensures that the crawled information has high consistency and reliability. Utilizing natural language processing technology to analyze the content of new source information can extract key information, perform sentiment analysis and topic classification, and provide users with a deeper understanding of the content. The constructed evaluation system considers multiple dimensions such as content quality, timeliness, authority, and audience feedback, and can objectively evaluate the quality and credibility of the source. The evaluation criteria and weights are dynamically adjusted based on user feedback, making the evaluation system more closely aligned with user needs and market changes. Based on user feedback and preferences, personalized source recommendations can be further provided to meet the needs of different users.

[0061] In this embodiment, the construction of the new information source filtering mechanism in step one specifically includes:

[0062] Identify the target field: Clarify the specific field of the new information source, such as artificial intelligence. Based on the characteristics of the new information source field, AI involves multiple disciplines such as computer science, mathematics, psychology, and neuroscience. The information sources are diverse and it has applications in multiple industries such as medicine, education, finance, and transportation. The information content is rich. Select relevant technology information keywords, such as: machine learning, deep learning, natural language processing, robotics, reinforcement learning, etc.

[0063] Data crawling: Based on the selected science and technology information keywords, use web crawling technology to send HTTP requests to Internet platforms to obtain the HTML source code of the web pages, parse the web page content, and extract links such as news reports, professional new information sources, and industry reports corresponding to various Internet platforms as seed links of potential information sources;

[0064] Set up a scheduled task to periodically run the crawler to access the target website. The crawler identifies and records the timestamp of the last crawling operation, compares the currently obtained content with the information recorded during the last crawl, and only crawls and collects the content added and updated since the last crawl.

[0065] In this embodiment, the latest content of the target website is monitored periodically by web crawling technology. An incremental crawling strategy is adopted, and the content updated since the last crawl is crawled, which improves crawling efficiency and reduces unnecessary resource consumption, thereby improving the efficiency and real-time performance of data acquisition.

[0066] Establish screening rules: Set the update frequency of information sources according to the timeliness requirements of the new information source field. For example, for news reports, daily updates can be required; for professional new information sources and industry reports, updates can be required within a certain period (such as weekly or monthly). The content quality of information sources should be comprehensively evaluated based on factors such as the source channel, author background, and reader evaluation, including the accuracy, authority, and depth of information. Information sources that match the needs and characteristics of the target audience should be selected. For example, if the target audience is professionals or researchers in the new information source field, then those sources that provide professional analysis and in-depth interpretation can be given priority.

[0067] In this embodiment, the analysis of new information source content in step two specifically includes:

[0068] Keyword extraction: Based on the extracted metadata, the title of the main content is identified, a summary of the main content is drawn, and target keywords are extracted from the main content, specifically including:

[0069] From the summary of the main content, continuous word combinations are extracted based on contextual relationships. The extracted word combinations are then concatenated according to grammatical rules and semantics to obtain concatenated sentences.

[0070] The concatenated sentences are tagged with parts of speech, and the sentences are cleaned based on part-of-speech statistical features and domain-specific terminology to obtain a keyword set;

[0071] Each word in the keyword set is converted into a corresponding word vector. The distance between each word vector and its corresponding standard word vector is calculated. The keyword corresponding to the word vector with the smallest distance is selected as the target keyword.

[0072] Natural Language Processing Analysis: Based on sentiment analysis technology, the sentiment tendency of new information sources is determined. Based on target keywords, the new information sources are classified into themes. At the same time, information clusters with the same theme are generated by integrating the content similarity between new information sources.

[0073] Content quality assessment: Based on the extracted metadata information, the author, publishing organization information and publication date are identified to detect whether the new source information is original, and it is compared with authoritative information to determine the accuracy of the new source information.

[0074] Source release pattern analysis: Analyze the release time of information sources to see if there are peak release periods, calculate the release frequency of information sources, and evaluate their activity level based on the calculation results;

[0075] Trend Analysis: Analyze the repetition intensity of titles and keywords in the main content to identify hot topics. Based on historical data, predict potential new news source trends and hot topics in the future. Historical data includes: the number and type of articles, reports, and news related to new news sources published in the past period; the historical search frequency and trends of keywords related to new news sources; changes in the frequency of keywords appearing in information; the changing trends of sentiment analysis results of past new news source information over time; changes in the popularity and attention of various topics over time; and analyze the breadth and depth of news source reporting content, including the diversity of topics involved and the geographical coverage.

[0076] In this embodiment, in-depth analysis of content can provide users with more personalized information services, clear topic classification and sentiment judgment, improve the user's reading experience, provide users with high-quality and highly relevant information, support users in making more informed decisions, and encourage information sources to provide better content through evaluation and feedback, promote the healthy development of the entire content ecosystem, and improve the transparency of information and reduce the problem of information asymmetry through comprehensive evaluation of information sources and content.

[0077] In this embodiment, generating information clusters on the same topic further includes:

[0078] Analyze the characteristics of each information cluster, such as the number of information items and user interaction, and determine one or more ranking factors associated with the topic of the information cluster based on the information cluster characteristics, such as the number of reads, the number of comments, and the number of shares.

[0079] Conduct content similarity analysis between information clusters to identify common keywords and themes in different information clusters, and construct an association network of information clusters based on the similarity analysis results;

[0080] Weights are assigned to each ranking factor, and a comprehensive score is calculated for each information cluster based on the ranking factor and weight. The corresponding information clusters are ranked according to the comprehensive score, and the degree of correlation between each information cluster and other topics is marked next to each information cluster. The ranking and correlation relationships are visualized using tools such as charts and heatmaps.

[0081] In this embodiment, a large amount of information is systematically organized by clustering and analyzing information clusters, which improves the efficiency of information processing. Users can quickly locate information clusters on relevant topics according to their interests, which improves the convenience of information retrieval. The value of information clusters is quantified by ranking factors and comprehensive scores, which helps users identify high-quality information. The construction of association networks reveals the connections between different topics, which enhances users' understanding of the overall picture of information. Based on the characteristics and ranking of information clusters, more accurate content recommendations can be provided to users.

[0082] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A new method for discovering and evaluating information sources for internet technology information, characterized in that, Includes the following steps: Step 1: Construct a new information source discovery mechanism: Determine the preset Internet technology information keywords and target fields, crawl information from Internet platforms in the target fields based on web crawling technology, obtain seed links of potential information sources, and set information source filtering rules; Step 2: Acquisition of new information sources: Automatically capture relevant new information source content based on the constructed new information source discovery mechanism, preprocess the captured new information source content, establish a new information source information database, and store the captured new information source content; Step 3: New Source Content Analysis: Extract key information from the new source information and analyze the new source information content using natural language processing technology; Step Four: Evaluation of New Information Sources: Based on the analysis results of the new information source content, an evaluation system is constructed. The information sources are scored according to this system, and then ranked according to their scores. This also includes: Establish a user feedback mechanism, collect user feedback data on the information source evaluation system at preset time intervals, and classify the collected feedback data; Identify positive and negative sentiments in user feedback data and adjust evaluation criteria and weights based on user feedback; Step 5: Output Results: Based on the scoring results, the information sources are divided into different credit levels, and then the information sources are arranged from high to low according to the credit level; The analysis of new information source content in step three specifically includes: Keyword extraction: Based on the extracted metadata information, identify the title of the main content, summarize the main content, and extract the target keywords from the main content; Natural Language Processing Analysis: Based on sentiment analysis technology, the sentiment tendency of new information sources is determined. Based on target keywords, the new information sources are classified into themes. At the same time, information clusters with the same theme are generated by integrating the content similarity between new information sources. Content quality assessment: Based on the extracted metadata information, the author, publishing organization information and publication date are identified to detect whether the new source information content is original, and it is compared with authoritative information to determine the accuracy of the new source information. Source release pattern analysis: Analyze the release time of information sources to see if there are peak release periods, calculate the release frequency of information sources, and evaluate the activity level based on the calculation results; Trend Analysis: Identify trending topics by analyzing the repetition intensity of titles and keywords in the main content, and predict potential new news source trends and hot topics based on historical data; Generating information clusters on the same topic also includes: Analyze the characteristics of each information cluster, and determine one or more ranking factors associated with information clusters on the same topic based on the information cluster characteristics; Conduct content similarity analysis between information clusters to identify common keywords and themes in different information clusters, and construct an association network of information clusters based on the similarity analysis results; A weight is assigned to each ranking factor, and a comprehensive score is calculated for each information cluster based on the ranking factor and weight. The corresponding information clusters are ranked according to the comprehensive score, and the degree of correlation between each information cluster and other topics is marked next to each information cluster.

2. The method for discovering and evaluating new sources of information for Internet technology information as described in claim 1, characterized in that, The construction of the new information source discovery mechanism in step one is specifically as follows: Identify the target area: Define the specific area of ​​the new information source and select relevant scientific and technological information keywords based on the characteristics of the new information source area; Data crawling: Based on the selected science and technology information keywords, use web crawling technology to send HTTP requests to Internet platforms to obtain the HTML source code of the web pages, parse the web page content, and extract the links corresponding to each Internet platform as seed links of potential information sources. Establish screening rules: Set the update frequency of information sources according to the timeliness requirements of new information source fields, comprehensively evaluate the content quality of information sources based on the quality influencing factors of information sources, and screen information sources that match the needs and characteristics of the target audience.

3. The method for discovering and evaluating new sources of information for Internet technology information as described in claim 2, characterized in that, The data crawling also includes: setting up a scheduled task, running a crawler program periodically to access the target website based on the set scheduled task, the crawler program identifying and recording the timestamp of the last crawling operation, comparing the currently obtained content with the information recorded during the last crawling, and only crawling and collecting the content added and updated since the last crawling.

4. The method for discovering and evaluating new sources of information for Internet technology information as described in claim 3, characterized in that, Step two involves preprocessing the captured information, specifically including: Obtain and identify HTML tags in the webpage, obtain the information attributes of the HTML tags, divide the content data referenced by the HTML tags according to the information attributes, and obtain a content data distribution map; At the same time, it identifies and removes advertising content from web pages based on the advertising recognition database; Based on the content data distribution map, the main content area in the page is identified, the metadata information of the main content is extracted, and the extracted information content and its metadata are processed into data structure.

5. The method for discovering and evaluating new sources of information for Internet technology information as described in claim 4, characterized in that, The new information source database includes a classification storage and identification model for information content tables, metadata tables, information source evaluation tables, and user feedback tables. The acquired data is classified and stored based on the classification storage and identification model.

6. The method for discovering and evaluating new sources of information for Internet technology information as described in claim 5, characterized in that, The extraction of target keywords from the main content specifically includes: From the summary of the main content, continuous word combinations are extracted based on contextual relationships. The extracted word combinations are then concatenated according to grammatical rules and semantics to obtain concatenated sentences. The concatenated sentences are tagged with parts of speech, and the sentences are cleaned based on part-of-speech statistical features and domain-specific terminology to obtain a keyword set; Each word in the keyword set is converted into a corresponding word vector. The distance between each word vector and its corresponding standard word vector is calculated. The keyword corresponding to the word vector with the smallest distance is selected as the target keyword.

7. The method for discovering and evaluating new sources of information for Internet technology information as described in claim 6, characterized in that, The output of the results in step five specifically includes: Based on the scoring results, a clear correspondence between the scoring score range and the credit rating is established, and the information sources are classified into the corresponding credit ratings based on the corresponding keywords. Information sources are sorted according to credit rating and score. For information sources with the same credit rating, a secondary sort is performed based on the specific score.

Images