A public opinion source tracking system based on multi-path retrieval

Abstract

The present invention provides a public opinion source tracing system based on multi-path retrieval, which relates to the field of electronic digital data processing, and comprises a target parsing module, an information retrieval module, a public opinion judgment module and a source tracing control module. The target parsing module is used to parse and process a target text, the information retrieval module establishes multiple retrieval paths based on the parsed content and performs information retrieval, the public opinion judgment module is used to perform public opinion relevance judgment on the retrieved information to filter out the public opinion information, and the source tracing control module is used to analyze the public opinion information, control retrieval parameters and determine the source of the public opinion. The system performs text retrieval by establishing multiple keyword-based retrieval paths and analyzing and processing the retrieved files, so as to effectively find the source information of the entire public opinion event and better manage the public opinion.

Description

Technical Field

[0001] The present invention relates to the field of electronic digital data processing, and in particular to a public opinion source tracing system based on multi-path retrieval. Background Art

[0002] With the rapid development of the Internet and social media, the speed of information dissemination has accelerated unprecedentedly. Public opinion has become one of the important factors affecting social stability and economic development. When major events occur, the speed and scope of public opinion dissemination often show an exponential growth trend. Therefore, timely and accurate tracking of the source of public opinion and understanding its dissemination path are of great significance for government departments, enterprises and social organizations to formulate response strategies, avoid crises and maintain social stability.

[0003] The foregoing discussion of the background art is intended only to facilitate an understanding of the present invention. This discussion does not acknowledge or admit that any of the material referred to is part of the common general knowledge.

[0004] Now many public opinion tracing systems have been developed. After a lot of searching and reference, it is found that the existing public opinion tracing systems are such as the system disclosed by the publication number CN117150104B. These systems generally include a basic path propagation data extraction module, a public opinion propagation status preliminary verification module, a public opinion propagation path traceability tracking module, a public opinion path source analysis confirmation module and a public opinion path source feedback terminal. The public opinion propagation status is initially verified according to the opinions of the users participating in the target public opinion topic and each derivative sub-public opinion topic, and the propagation path is screened. The public opinion tracking analysis is performed on the screened propagation path to confirm the public opinion propagation source path and the public opinion source propagation user. However, in the process of tracing, due to the development and changes of public opinion events, the system is prone to miss the real source information of the event, and the accuracy of tracing needs to be improved. Summary of the invention

[0005] The purpose of the present invention is to propose a public opinion source tracing system based on multi-path retrieval to address the existing deficiencies.

[0006] The present invention adopts the following technical solution:

[0007] A public opinion source tracing system based on multi-path retrieval, including a target parsing module, an information retrieval module, a public opinion judgment module and a tracing control module;

[0008] The target parsing module is used to parse the target text, the information retrieval module establishes multiple retrieval paths based on the parsed content and performs information retrieval, the public opinion judgment module is used to judge the public opinion relevance of the retrieved information and filter out the public opinion information, and the source tracing control module is used to analyze the public opinion information, control the retrieval parameters and determine the source of the public opinion;

[0009] The target parsing module includes a target input unit and a keyword extraction unit, wherein the target input unit is used to obtain event information as target text, and the keyword extraction unit is used to extract keyword information;

[0010] The information retrieval module includes a path construction unit, a text retrieval unit and a path correction unit. The path construction unit constructs a retrieval path based on keyword information, the text retrieval unit retrieves information in the network based on keywords, and the path correction unit is used to delete the retrieval path;

[0011] The public opinion judgment module includes a key information comparison unit and a conflict judgment unit. The key information comparison unit is used to compare the keywords in the search information with the keywords in the target text, and the conflict judgment unit is used to perform conflict analysis on the comparison results and judge whether to retain the public opinion information;

[0012] The traceability control module includes a path parsing unit, a public opinion parsing unit and a time parsing unit. The path parsing unit is used to analyze the amount of public opinion information in each search path and determine the search path that needs to be deleted. The public opinion parsing unit is used to analyze the content of all public opinion information and determine the search path that needs to be added. The time parsing unit is used to analyze the time distribution of public opinion information and determine whether there is a public opinion source.

[0013] Further, the path construction unit includes a storage space application processor, an information sorting processor and a space management processor, wherein the storage space application processor is used to apply for a storage space for each received keyword, the information sorting processor is used to sort the information in the storage space according to the time sequence, and the space management processor is used to manage the storage space;

[0014] Further, the key information comparison unit includes a path identification processor, an information conversion processor and a comparison classification processor, wherein the path identification processor is used to identify the storage space and obtain the corresponding keyword information, the information conversion processor is used to obtain the information text from the storage space and convert it into keywords, and the comparison classification processor is used to compare and classify the two groups of keywords;

[0015] Further, the path parsing unit includes a path calculation processor and a first instruction sending processor, the path calculation processor is used to calculate the retention index of the search path, and the first instruction sending processor is used to send a path deletion instruction;

[0016] The path calculation processor calculates the retention index Vr of the search path according to the following formula:

[0017] ;

[0018] Among them, n r is the number of public opinion texts in the storage space, n d The number of retrieved texts deleted from the storage space;

[0019] When the retention index is less than the retention threshold, the corresponding search path needs to be deleted;

[0020] Furthermore, the time analysis unit includes an interval statistics processor and a source judgment processor, wherein the interval statistics processor is used to count the time intervals between the release of public opinion texts in each storage space, and the source calculation processor is used to calculate the source index of each search path;

[0021] The source calculation processor calculates the source index Vs of the search path according to the following formula:

[0022] ;

[0023] Among them, M is the number of public opinion texts in the search path, T i It represents the time interval between the ith public opinion text and the i+1th public opinion text, and Td is the standard ratio value;

[0024] When the source index is less than the source threshold, the first public opinion text in the storage space is determined to be the phased source file of the corresponding search path;

[0025] When the number of search paths to which the earliest stage source file published in all search paths belongs reaches a comprehensive threshold, the stage source file is determined to be the source file of the entire public opinion.

[0026] The beneficial effects achieved by the present invention are:

[0027] This system establishes a search path through keywords, analyzes the search text in the search path, and continuously changes keywords upward, so as not to miss the real source event information. During the search process, time segment search is performed to make the search process controllable, and at the same time make the retrieved text information more consistent, which is conducive to more accurate keyword changes.

[0028] To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are only for reference and description and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a schematic diagram of the overall structural framework of the present invention;

[0030] Figure 2 This is a schematic diagram of the target analysis module of the present invention;

[0031] Figure 3 This is a schematic diagram of the information retrieval module of the present invention;

[0032] Figure 4 This is a schematic diagram of the public opinion judgment module of the present invention;

[0033] Figure 5 This is a schematic diagram of the traceability control module of the present invention. DETAILED DESCRIPTION

[0034] The following is an explanation of the embodiments of the present invention through specific embodiments. Those skilled in the art can understand the advantages and effects of the present invention from the contents disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and the details in this specification can also be modified and changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted according to actual sizes. It is stated in advance. The following embodiments will further explain the relevant technical contents of the present invention in detail, but the disclosed contents are not intended to limit the scope of protection of the present invention.

[0035] Embodiment 1: This embodiment provides a public opinion tracing system based on multi-path retrieval, combined with Figure 1 , including target parsing module, information retrieval module, public opinion judgment module and traceability control module;

[0036] The target parsing module is used to parse the target text, the information retrieval module establishes multiple retrieval paths based on the parsed content and performs information retrieval, the public opinion judgment module is used to judge the public opinion relevance of the retrieved information and filter out the public opinion information, and the source tracing control module is used to analyze the public opinion information, control the retrieval parameters and determine the source of the public opinion;

[0037] The target parsing module includes a target input unit and a keyword extraction unit, wherein the target input unit is used to obtain event information as target text, and the keyword extraction unit is used to extract keyword information;

[0038] The information retrieval module includes a path construction unit, a text retrieval unit and a path correction unit. The path construction unit constructs a retrieval path based on keyword information, the text retrieval unit retrieves information in the network based on keywords, and the path correction unit is used to delete the retrieval path;

[0039] The public opinion judgment module includes a key information comparison unit and a conflict judgment unit. The key information comparison unit is used to compare the keywords in the search information with the keywords in the target text, and the conflict judgment unit is used to perform conflict analysis on the comparison results and judge whether to retain the public opinion information;

[0040] The traceability control module includes a path parsing unit, a public opinion parsing unit and a time parsing unit. The path parsing unit is used to analyze the amount of public opinion information in each search path and determine the search path that needs to be deleted. The public opinion parsing unit is used to analyze the content of all public opinion information and determine the search path that needs to be added. The time parsing unit is used to analyze the time distribution of public opinion information and determine whether there is a public opinion source.

[0041] The path construction unit includes a storage space application processor, an information sorting processor and a space management processor, wherein the storage space application processor is used to apply for a storage space for each received keyword, the information sorting processor is used to sort the information in the storage space according to the time sequence, and the space management processor is used to manage the storage space;

[0042] The key information comparison unit includes a path identification processor, an information conversion processor and a comparison classification processor, wherein the path identification processor is used to identify the storage space and obtain the corresponding keyword information, the information conversion processor is used to obtain the information text from the storage space and convert it into keywords, and the comparison classification processor is used to compare and classify the two groups of keywords;

[0043] The path parsing unit includes a path calculation processor and a first instruction sending processor, the path calculation processor is used to calculate the retention index of the search path, and the first instruction sending processor is used to send a path deletion instruction;

[0044] The path calculation processor calculates the retention index Vr of the search path according to the following formula:

[0045] ;

[0046] Among them, n r is the number of public opinion texts in the storage space, n d The number of retrieved texts deleted from the storage space;

[0047] When the retention index is less than the retention threshold, the corresponding search path needs to be deleted;

[0048] The time analysis unit includes an interval statistics processor and a source judgment processor, wherein the interval statistics processor is used to count the time intervals between the release of public opinion texts in each storage space, and the source calculation processor is used to calculate the source index of each search path;

[0049] The source calculation processor calculates the source index Vs of the search path according to the following formula:

[0050] ;

[0051] Among them, M is the number of public opinion texts in the search path, T i It represents the time interval between the ith public opinion text and the i+1th public opinion text, and Td is the standard ratio value;

[0052] When the source index is less than the source threshold, the first public opinion text in the storage space is determined to be the phased source file of the corresponding search path;

[0053] When the number of search paths to which the earliest stage source file published in all search paths belongs reaches a comprehensive threshold, the stage source file is determined to be the source file of the entire public opinion.

[0054] Embodiment 2: This embodiment includes all the contents of Embodiment 1, and provides a public opinion source tracing system based on multi-path retrieval, including a target parsing module, an information retrieval module, a public opinion judgment module and a source tracing control module;

[0055] The target parsing module is used to parse the target text, the information retrieval module establishes multiple retrieval paths based on the parsed content and performs information retrieval, the public opinion judgment module is used to judge the public opinion relevance of the retrieved information and filter out the public opinion information, and the source tracing control module is used to analyze the public opinion information, control the retrieval parameters and determine the source of the public opinion;

[0056] Combination Figure 2 , the target parsing module includes a target input unit and a keyword extraction unit, the target input unit is used to obtain event information as the target text, and the key extraction unit is used to extract keyword information;

[0057] Combination Figure 3 The information retrieval module includes a path construction unit, a text retrieval unit and a path correction unit, wherein the path construction unit constructs a retrieval path based on keyword information, the text retrieval unit retrieves information in the network based on keywords, and the path correction unit is used to delete the retrieval path;

[0058] Combination Figure 4 The public opinion judgment module includes a key information comparison unit and a conflict judgment unit. The key information comparison unit is used to compare the keywords in the search information with the keywords in the target text, and the conflict judgment unit is used to perform conflict analysis on the comparison results and judge whether to retain the public opinion information;

[0059] Combination Figure 5The traceability control module includes a path parsing unit, a public opinion parsing unit and a time parsing unit. The path parsing unit is used to analyze the amount of public opinion information in each search path and determine the search path that needs to be deleted. The public opinion parsing unit is used to analyze the content of all public opinion information and determine the search path that needs to be added. The time parsing unit is used to analyze the time distribution of public opinion information and determine whether there is a public opinion source.

[0060] The process of public opinion tracing in this system includes the following steps:

[0061] S1, the target input unit receives the link information and obtains the corresponding target text;

[0062] S2, the keyword extraction unit extracts keywords from the target text;

[0063] S3, a path building unit builds a search path based on each keyword;

[0064] S4, the text retrieval unit retrieves and collects information from the network based on keywords;

[0065] S5, the public opinion judgment module judges the collected information and retains the public opinion information;

[0066] S6, the path parsing unit analyzes and obtains the search path that needs to be deleted and sends an instruction to the path correction unit;

[0067] S7, the public opinion analysis unit analyzes and obtains new keywords and sends instructions to the path correction unit;

[0068] S8, the path correction unit deletes the search path based on the received instruction;

[0069] S9, the time analysis unit analyzes and determines whether there is a public opinion source. If yes, the tracing process ends. If no, the search time period is updated and sent to the text search unit, and the process returns to step S4;

[0070] After updating the search time period, the public opinion information in the search path is also reset;

[0071] The target input unit includes a link receiving processor, a network access processor and a text copying processor, wherein the link receiving processor is used to receive link information, the network access processor is used to access the network target pointed to by the link information, and the text copying processor is used to copy and store text information in the network target;

[0072] The keyword extraction unit includes a text segmentation processor, a comparison extraction processor and a feedback control processor, wherein the text segmentation processor is used to segment text information into a plurality of short sentences, the comparison extraction processor is used to compare the short sentences with the keyword library and extract the keywords, and the feedback control processor is used for the feedback object of the keywords;

[0073] When the keyword extraction unit obtains text information from the text copy processor, the extracted keywords are sent to the information retrieval module; when the keyword extraction unit obtains text information sent by the public opinion judgment module, the extracted keywords are sent to the key information comparison unit;

[0074] The path construction unit includes a storage space application processor, an information sorting processor and a space management processor, wherein the storage space application processor is used to apply for a storage space for each received keyword, the information sorting processor is used to sort the information in the storage space according to the time sequence, and the space management processor is used to manage the storage space;

[0075] The text retrieval unit includes a space mapping processor and a network retrieval processor, wherein the space mapping processor is used to identify the storage space and map the corresponding keywords, and record the mapping relationship between the storage address of the storage space and the keywords; the network retrieval processor is used to retrieve network information containing the keywords in the network, and the retrieved network information is saved in the corresponding storage space based on the mapping relationship;

[0076] The text stored in the storage space is called the retrieved text;

[0077] The network search processor sets a search time interval during the search process and only searches for network information within the search time interval;

[0078] The path correction unit includes an instruction receiving processor and an instruction interpretation processor, wherein the instruction receiving processor is used to receive instruction information, and the instruction interpretation processor is used to interpret the instruction and send the interpretation result to the space management processor;

[0079] The key information comparison unit includes a path identification processor, an information conversion processor and a comparison classification processor, wherein the path identification processor is used to identify the storage space and obtain the corresponding keyword information, the information conversion processor is used to obtain the information text from the storage space and convert it into keywords, and the comparison classification processor is used to compare and classify the two groups of keywords;

[0080] The information conversion processor sends the acquired information text to the keyword extraction unit and obtains the corresponding keyword;

[0081] The comparison and classification processor classifies keywords into basic keywords, path keywords and actual keywords, wherein the basic keywords refer to keywords that are present in both the search path and the search text, the path keywords refer to keywords that are present in the search path but not in the search text, and the actual keywords refer to keywords that are not present in the search path but in the search text;

[0082] The conflict judgment unit includes an initial judgment processor and a conflict calculation processor, wherein the initial judgment processor performs a preliminary judgment based on the number of basic keywords, and the conflict calculation processor is used to calculate the conflict between the path keyword and the time keyword;

[0083] When the number of basic keywords is less than the basic threshold, it is directly determined that the retrieved text does not belong to public opinion information;

[0084] The conflict calculation processor calculates the conflict index P between the path keyword and the actual keyword according to the following formula:

[0085] ;

[0086] Where n is the number of path keywords, m is the number of time keywords, and c(i, j) represents the conflict value between the i-th path keyword and the j-th time keyword;

[0087] The conflict values ​​between the various keywords are pre-set in the conflict calculation processor;

[0088] When the conflict index is greater than the conflict threshold, it is determined that the retrieved text does not belong to public opinion information;

[0089] The searched text judged as public opinion information is called public opinion text, and the searched text judged as non-public opinion information is deleted from the storage space;

[0090] The path parsing unit includes a path calculation processor and a first instruction sending processor, the path calculation processor is used to calculate the retention index of the search path, and the first instruction sending processor is used to send a path deletion instruction;

[0091] The path calculation processor calculates the retention index Vr of the search path according to the following formula:

[0092] ;

[0093] Among them, n r is the number of public opinion texts in the storage space, n d The number of retrieved texts deleted from the storage space;

[0094] When the retention index is less than the retention threshold, the corresponding search path needs to be deleted;

[0095] The public opinion analysis unit includes a keyword receiving processor and an adding judgment processor, wherein the keyword receiving processor is used to receive actual keyword information of the public opinion text in the public opinion judgment module, and the adding judgment processor is used to judge whether to establish a corresponding search path according to the time keyword;

[0096] The adding judgment processor calculates the adding index Vc of the actual keyword according to the following formula:

[0097] ;

[0098] Where N is the number of search paths, a i is the proportion of public opinion texts containing the actual keyword in the i-th storage space;

[0099] When the addition index is greater than the addition threshold, the search path corresponding to the actual keyword needs to be added;

[0100] The time analysis unit includes an interval statistics processor and a source judgment processor, wherein the interval statistics processor is used to count the time intervals between the release of public opinion texts in each storage space, and the source calculation processor is used to calculate the source index of each search path;

[0101] The source calculation processor calculates the source index Vs of the search path according to the following formula:

[0102] ;

[0103] Among them, M is the number of public opinion texts in the search path, T i It represents the time interval between the ith public opinion text and the i+1th public opinion text, and Td is the standard ratio value;

[0104] When the source index is less than the source threshold, the first public opinion text in the storage space is determined to be the phased source file of the corresponding search path;

[0105] When the number of search paths to which the earliest interim source file published in all search paths belongs reaches the comprehensive threshold, the interim source file is determined to be the source file of the entire public opinion;

[0106] The i and j appearing in the above text are ordinal numbers used to indicate sequence numbers.

[0107] The contents disclosed above are only preferred feasible embodiments of the present invention, and do not limit the protection scope of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention specification and drawings are included in the protection scope of the present invention. In addition, the elements therein can be updated as technology develops.

Claims

1. A public opinion source tracking system based on multi-path retrieval, characterized in that: It includes target analysis module, information retrieval module, public opinion judgment module and source tracing control module; The target parsing module is used to parse the target text, the information retrieval module establishes multiple retrieval paths based on the parsed content and performs information retrieval, the public opinion judgment module is used to judge the public opinion relevance of the retrieved information and filter out the public opinion information, and the source tracing control module is used to analyze the public opinion information, control the retrieval parameters and determine the source of the public opinion; The target parsing module includes a target input unit and a keyword extraction unit, wherein the target input unit is used to obtain event information as target text, and the keyword extraction unit is used to extract keyword information; The information retrieval module includes a path construction unit, a text retrieval unit and a path correction unit. The path construction unit constructs a retrieval path based on keyword information, the text retrieval unit retrieves information in the network based on keywords, and the path correction unit is used to delete the retrieval path; The public opinion judgment module includes a key information comparison unit and a conflict judgment unit. The key information comparison unit is used to compare the keywords in the search information with the keywords in the target text, and the conflict judgment unit is used to perform conflict analysis on the comparison results and judge whether to retain the public opinion information; The traceability control module includes a path parsing unit, a public opinion parsing unit and a time parsing unit. The path parsing unit is used to analyze the amount of public opinion information in each search path and determine the search path that needs to be deleted. The public opinion parsing unit is used to analyze the content of all public opinion information and determine the search path that needs to be added. The time parsing unit is used to analyze the time distribution of public opinion information and determine whether there is a public opinion source; the path parsing unit includes a path calculation processor and a first instruction sending processor. The path calculation processor is used to calculate the retention index of the search path, and the first instruction sending processor is used to send a path deletion instruction; the path calculation processor calculates the retention index Vr of the search path according to the following formula: ； Among them, n r is the number of public opinion texts in the storage space, n d The number of retrieved texts deleted from the storage space; When the retention index is less than the retention threshold, the corresponding search path needs to be deleted.

2. A public opinion source tracing system based on multi-path retrieval as claimed in claim 1, characterized in that: The path construction unit includes a storage space application processor, an information sorting processor and a space management processor. The storage space application processor is used to apply for a storage space for each received keyword, the information sorting processor is used to sort the information in the storage space according to time sequence, and the space management processor is used to manage the storage space.

3. A public opinion source tracing system based on multi-path retrieval as claimed in claim 2, characterized in that: The key information comparison unit includes a path identification processor, an information conversion processor and a comparison classification processor. The path identification processor is used to identify the storage space and obtain the corresponding keyword information. The information conversion processor is used to obtain the information text from the storage space and convert it into keywords. The comparison classification processor is used to compare and classify two groups of keywords.

4. A public opinion source tracing system based on multi-path retrieval as claimed in claim 3, characterized in that: The time parsing unit includes an interval statistics processor and a source judgment processor, wherein the interval statistics processor is used to count the time intervals between the release of public opinion texts in each storage space, and the source judgment processor is used to calculate the source index of each search path; The source determination processor calculates the source index Vs of the search path according to the following formula: ； Among them, M is the number of public opinion texts in the search path, T i It represents the time interval between the ith public opinion text and the i+1th public opinion text, and Td is the standard ratio value; When the source index is less than the source threshold, the first public opinion text in the storage space is determined to be the phased source file of the corresponding search path; When the number of search paths to which the earliest stage source file published in all search paths belongs reaches a comprehensive threshold, the stage source file is determined to be the source file of the entire public opinion.

Images