A smart travel big data public opinion monitoring and analyzing system and method

Abstract

The application discloses a smart travel and tourism big data public opinion monitoring and analyzing system and method, relates to the technical field of public opinion monitoring, and solves the technical problem that the prior art does not consider auditing the authenticity of obtained public opinion information, thereby affecting the selection and decision of subsequent tourists on scenic spots, and causing negative influence on the economic income and reputation of the scenic spots; the application collects public opinion information from multiple communication media to construct a public opinion information database; sets keywords and screens public opinion information from the public opinion information database according to the keywords; constructs and trains a public opinion type judgment model; determines the type of public opinion information in the query time range according to the public opinion type judgment model; and constructs a public opinion correction factor to correct the type of public opinion information; the application verifies the authenticity of public opinion information, helps to improve the correct understanding of tourists on scenic spots, and promotes the economic development of the scenic spots to a great extent.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of Internet, and relates to public opinion monitoring technology, in particular to a smart cultural and tourism big data public opinion monitoring and analysis system and method. BACKGROUND

[0002] The smart cultural and tourism big data public opinion monitoring and analysis system not only has a monitoring function, but also can provide targeted public opinion guidance and response suggestions according to the public opinion analysis results. This helps the cultural and tourism department or scenic spot to take timely measures to guide public opinion and reduce the influence of negative public opinion, and improve the public's trust and satisfaction with the cultural and tourism industry. The construction and application of the smart cultural and tourism big data public opinion monitoring and analysis system is an important embodiment of the intelligent development of the cultural and tourism industry. By introducing big data technology and intelligent algorithms, the system can realize automatic and intelligent processing and analysis of public opinion information, improve the efficiency and quality of public opinion monitoring and response. This will help promote the development of the cultural and tourism industry in a more intelligent and efficient direction.

[0003] The prior art (CN116955621A) discloses a monitoring method suitable for tourism public opinion, including the following steps: determining the target to be monitored and the data source to be collected; using network crawler technology to capture text data related to the monitoring target in the data source, and filtering and preprocessing the captured data to remove irrelevant information; performing data mining and public opinion analysis on the preprocessed text data, extracting key information, and classifying and summarizing the key information to mark the sentiment of the text; visualizing the analysis results and generating corresponding reports and analysis results; the prior art performs data mining and public opinion analysis on text data, without considering the accuracy and authenticity of the obtained public opinion information, and also does not actively guide public opinion and correct false public opinion, thereby affecting the selection and decision of subsequent tourists on the scenic spot, and causing negative impact on the economic income and reputation of the scenic spot.

[0004] The present application provides a smart cultural and tourism big data public opinion monitoring and analysis system and method to solve the above technical problems. SUMMARY

[0005] The present application aims to solve at least one of the technical problems existing in the prior art; for this purpose, the present application provides a smart cultural and tourism big data public opinion monitoring and analysis system to solve the technical problem that the prior art does not consider the accuracy and authenticity of the obtained public opinion information, and also does not actively guide public opinion and correct false public opinion, thereby affecting the selection and decision of subsequent tourists on the scenic spot, and causing negative impact on the economic income and reputation of the scenic spot.

[0006] To achieve the above objectives, a first aspect of the present invention provides a smart cultural tourism big data public opinion monitoring and analysis system, comprising: a public opinion collection module and a public opinion analysis module connected thereto;

[0007] Public opinion collection module: Collects public opinion information from multiple media outlets to build a public opinion information database; sets keywords and filters several pieces of public opinion information from the database based on the keywords; the several pieces of public opinion information are sorted in chronological order of occurrence; wherein, the keywords include: scenic spot name, custom query terms; public opinion information includes: occurrence time, end time, scenic spot name, and custom query terms; and the occurrence time and end time are determined based on the popularity and click-through rate or search rate of the platform where the public opinion is located;

[0008] Public opinion analysis module: Constructs and trains a public opinion type judgment model; determines the type of public opinion information within the query time range based on the public opinion type judgment model; and,

[0009] A public opinion correction factor is constructed to correct the type of public opinion information; the types of public opinion information include: positive, negative and neutral.

[0010] Preferably, the system includes: a public opinion feedback module connected to the public opinion analysis module: used to push types of public opinion information to users, and used to update the types of public opinion information.

[0011] Preferably, the step of setting keywords and filtering several pieces of public opinion information from the public opinion information database based on the keywords includes:

[0012] S110: Mark the scenic area name as the first keyword A, and mark the custom query term as the second keyword Bj; where Bj represents the j-th second keyword; j=1, 2, 3, ..., m; m is a positive integer;

[0013] S120: Set the query interval duration for public opinion information;

[0014] S130: Set the format of the j-th token to A+ Within the query time range, all historical public opinion information is sequentially searched from the public opinion information database to obtain historical public opinion information corresponding to different tags, and the number of historical public opinion information corresponding to different tags is counted and marked as such. .

[0015] It should be noted that the public opinion information database includes historical public opinion information as well as public opinion information within a set query period.

[0016] This invention filters search results based on relevance to ensure that the obtained public opinion information is closely related to the keywords; keyword filtering helps to select public opinion information based on keywords, giving priority to public opinion information with greater influence and attention; timeliness filtering helps to ensure that the selected public opinion information is up-to-date, reflecting the current public opinion dynamics and trends.

[0017] Preferably, the duration of the set query interval for public opinion information includes:

[0018] Public opinion information about the scenic area was retrieved from the public opinion database. The number of times (k) negative public opinion events occurred at the scenic area within the past two years was statistically analyzed, and the average duration of each historical public opinion event was marked as [missing information]. ; Obtain the revenue of the scenic area each month when there are no public opinion incidents; where k = 0, 1, 2, ..., n; n ∈ [0, +∞];

[0019] Determine whether the most recent public opinion event caused the scenic area's monthly revenue to fall below a preset low-revenue threshold; if so, the public opinion event is marked as having an impact on the local scenic area's revenue. If SR=1, then the occurrence of public opinion will be marked as having no impact on the local scenic area's revenue; otherwise, SR=1. and set =0;

[0020] The duration of the end of several public opinion events is marked as follows: ;

[0021] Through formula The range influence factor was calculated.

[0022] Through formula The duration of the query interval is calculated.

[0023] It should be noted that the specific value of the low-income threshold is set by the local cultural and tourism department staff based on the monthly revenue of the local scenic spots.

[0024] This invention determines the query time range for public opinion information by taking into account the number of times negative public opinion information appears in the corresponding scenic spot in the past and the degree of influence of the past public opinion on the current public opinion, which is conducive to capturing and evaluating the type of current real-time public opinion more quickly.

[0025] Preferably, the construction and training of the public opinion type judgment model includes:

[0026] S210: Retrieve historical public opinion information corresponding to different markers, and obtain the types of historical public opinion information;

[0027] S220: Retrieve the public opinion type judgment model and mark the number of historical public opinion information corresponding to the marker j=1. ;Will The corresponding historical public opinion information is sequentially input into the public opinion type judgment model; it is judged whether the type of the historical public opinion information output by the public opinion type judgment model is consistent with the type of the marked historical public opinion information; if yes, the output result of the public opinion type judgment model is marked as accurate; if not, the output result of the public opinion type judgment model is marked as inaccurate; the number of public opinion information with accurate output results is counted and recorded as . Among them, the public opinion type judgment model is built based on an artificial intelligence model;

[0028] Through formula The accuracy rate of historical public opinion information when j=1 was calculated.

[0029] S230: Will , , , ..., Each corresponding historical public opinion information is sequentially input into the public opinion type judgment model. Step S220 is repeated to obtain the accuracy of historical public opinion information corresponding to different markers.

[0030] S240: Construct a rectangular coordinate system with the number of keywords as the x-axis and accuracy as the y-axis. Count the total number of curve slopes between adjacent points in the rectangular coordinate system, and the number of curves with a slope greater than or equal to 0 between adjacent points. Determine if the proportion of curves with slopes greater than or equal to 0 to the total number of curves exceeds a threshold. If yes, the training of the public opinion type judgment model is complete; otherwise, use the first item as... The number of historical public opinion information input into the public opinion type judgment model is increased sequentially with a tolerance of 2 until the proportion of the number of curves with a slope greater than or equal to 0 is greater than the proportion threshold, at which point the training process ends.

[0031] It should be noted that when the number of historical public opinion information corresponding to different markers is odd, the number of historical public opinion information input to the public opinion type judgment model is determined by rounding.

[0032] This invention provides a direct way to assess the accuracy of a public opinion type judgment model by inputting historical public opinion information into the model and comparing the consistency between the model's output type and the labeled type. This evaluation method is both direct and effective, helping to identify problems and shortcomings in the model in a timely manner. During model training, the difficulty and complexity of training can be flexibly controlled by adjusting the number of input public opinion information, which helps to gradually improve the model's performance until the accuracy requirements are met. By constructing a Cartesian coordinate system and statistically analyzing the slope of the curve between adjacent points, the changing trend of the model's performance can be visually reflected.

[0033] Preferably, the public opinion type judgment model is built based on an artificial intelligence model, including:

[0034] Obtain standard training data; integrate all historical public opinion information into a display word cloud using word cloud generation technology, and arrange the displayed words in the display word cloud in descending order of font size to form a judgment factor sequence; the standard training data includes standard input data consistent with the content attributes of the judgment factor sequence, and standard output data consistent with the content attributes of the judgment factor sequence; the standard output data represents the type of public opinion information; use the standard training data to train the artificial intelligence model;

[0035] The trained artificial intelligence model is labeled as a public opinion type judgment model; the artificial intelligence model includes convolutional neural network model or long short-term memory neural network model.

[0036] Preferably, determining the type of public opinion information within the query time range based on the public opinion type judgment model includes:

[0037] Retrieve public opinion information within the query time range and obtain a word cloud of the public opinion information within the query time range. Analyze the sentiment type of several displayed words in the word cloud based on a sentiment dictionary. The sentiment type includes: favorability score a, favorability score b, favorability score 0, favorability score -b, and favorability score -a. Count the total number of displayed words in the word cloud and the number of displayed words with a favorability score greater than 0 in each sentiment type. Divide the number of displayed words with a favorability score greater than 0 in each sentiment type by the total number of displayed words to obtain the favorability ratio; where a > b > 0.

[0038] Determine if the favorability rating is greater than 80%; if yes, mark the corresponding public opinion information as positive; if no, determine if the favorability rating is less than 20%; if yes, mark the corresponding public opinion information as negative.

[0039] If not, the corresponding public opinion information will be marked as neutral.

[0040] This invention analyzes the sentiment type of displayed words based on a sentiment dictionary, which can accurately determine the sentiment tendency of public opinion information. By statistically analyzing the number of all displayed words in the word cloud and the number of displayed words with a favorability score greater than 0 in the sentiment type, the overall sentiment tendency of public opinion information can be quantitatively assessed. This quantitative method helps to more objectively judge the nature of public opinion information. The type of public opinion information is determined based on the favorability score ratio, realizing intelligent classification of public opinion information. This classification method is both fast and accurate, which helps to quickly respond to and process different types of public opinion information.

[0041] Preferably, the construction of the public opinion correction factor to correct the type of public opinion information includes:

[0042] Retrieve neutral public opinion information and count the number of likes (DZL), reposts (ZFL), and comments corresponding to neutral public opinion information with a favorability ratio within the range of [20%, 40%] ∪ [60%, 80%]. Remove invalid comments from the comments. Mark the number of comments after removing invalid comments as PLZ. Invalid comments are defined as comments with fewer than three characters, duplicate comments, and comments whose content is entirely non-text. Mark the number of invalid comments as WXL.

[0043] Through formula The public opinion correction factor was calculated; among which, and All are correction factors;

[0044] If the public opinion correction factor is greater than 0, replace the marked neutral public opinion information with a positive one; otherwise, if the public opinion correction factor is less than 0, replace the marked neutral public opinion information with a negative one; otherwise, do not replace the marked neutral public opinion information.

[0045] This invention reduces the misjudgment rate of public opinion information and improves the accuracy and practicality of judging the type of public opinion information by further subdividing and correcting neutral public opinion information.

[0046] Preferably, the method for obtaining the correction coefficient includes:

[0047] Retrieve the number of bits e of the likes (DZL) and the number of bits w of the shares (ZFL).

[0048] The correction factor α is calculated as follows: α = ;

[0049] The correction factor β is calculated as follows: β = ; where the range of values ​​for e and w is [0, +∞).

[0050] To achieve the above objectives, a second aspect of the present invention provides a smart cultural tourism big data public opinion monitoring and analysis method, comprising:

[0051] A public opinion information database is constructed by collecting public opinion information from multiple media outlets; keywords are set and several pieces of public opinion information are selected from the database based on the keywords; the several pieces of public opinion information are sorted in chronological order of their occurrence; wherein, the keywords include: scenic spot name and custom query terms; the public opinion information includes: the occurrence time of the public opinion, the end time, the scenic spot name, and the custom query terms; and the occurrence time and end time are determined based on the popularity and click-through rate or search rate of the platform where the public opinion is located;

[0052] Construct and train a public opinion type judgment model; determine the type of public opinion information within the query time range based on the public opinion type judgment model; and,

[0053] A public opinion correction factor is constructed to correct the type of public opinion information; the types of public opinion information include: positive, negative and neutral.

[0054] Compared with the prior art, the beneficial effects of the present invention are:

[0055] 1. This invention ensures the comprehensiveness and diversity of public opinion information by collecting it from multiple media outlets, avoiding bias caused by a single source. By setting keywords to filter information, it is easier to accurately locate public opinion related to the topic, improving the accuracy of the information. By constructing and training a public opinion type judgment model, it enables rapid judgment of public opinion information types, greatly improving processing efficiency. The automated process reduces manual intervention, lowers labor costs, and improves processing speed and consistency. The public opinion type judgment model can be adjusted and optimized according to actual needs to adapt to different public opinion environments and requirements.

[0056] 2. For public opinion information judged to be neutral, this invention further narrows the range and constructs a public opinion correction factor to correct the type of public opinion information, thereby improving the accuracy and practicality of public opinion information type judgment; by classifying and correcting the type of public opinion information, it can provide decision-makers with clear and accurate public opinion information, helping them to make more informed decisions; by obtaining three types of public opinion information—positive, negative, and neutral—it helps tourists to comprehensively and objectively understand the attitudes and opinions of other members of the public on different topics. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a schematic diagram illustrating the module relationships included in this invention;

[0059] Figure 2 This is a schematic diagram illustrating the specific steps involved in obtaining public opinion information according to the present invention.

[0060] Figure 3 This is a schematic diagram illustrating the specific steps of training the public opinion type judgment model of the present invention;

[0061] Figure 4 This is a schematic diagram illustrating the specific steps involved in revising the public opinion information type according to the present invention;

[0062] Figure 5 This is a schematic diagram of the public opinion detection and analysis process of this invention. Detailed Implementation

[0063] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. 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.

[0064] Please see Figure 1 The first aspect of the present invention provides a smart cultural tourism big data public opinion monitoring and analysis system, including: a public opinion collection module, a public opinion analysis module, and a public opinion feedback module connected thereto;

[0065] Public opinion collection module: Collects public opinion information from multiple media outlets to build a public opinion information database; sets keywords and filters several pieces of public opinion information from the database based on the keywords; the several pieces of public opinion information are sorted in chronological order of occurrence; wherein, the keywords include: scenic spot name, custom query terms; public opinion information includes: occurrence time, end time, scenic spot name, and custom query terms; and the occurrence time and end time are determined based on the popularity and click-through rate or search rate of the platform where the public opinion is located;

[0066] Public opinion analysis module: Constructs and trains a public opinion type judgment model; determines the type of public opinion information within the query time range based on the public opinion type judgment model; and,

[0067] A public opinion correction factor is constructed to correct the type of public opinion information; the types of public opinion information include: positive, negative, and neutral.

[0068] Public opinion feedback module: used to push different types of public opinion information to users, and to update the types of public opinion information.

[0069] Please see Figure 2 Specific steps for obtaining public opinion information

[0070] S110: Mark the scenic area name as the first keyword A, and mark the custom query term as the second keyword Bj; where Bj represents the j-th second keyword; j=1, 2, 3, ..., m; m is a positive integer;

[0071] S120: Retrieve public opinion information about the scenic area from the public opinion information database, and count the number of times (k) negative public opinion occurred at the corresponding scenic area within the past two years. The average duration of each historical public opinion event is marked as [missing information]. ; Obtain the revenue of the scenic area each month when there are no public opinion incidents; where k = 0, 1, 2, ..., n; n ∈ [0, +∞];

[0072] Determine whether the most recent public opinion event caused the scenic area's monthly revenue to fall below a preset low-revenue threshold; if so, the public opinion event is marked as having an impact on the local scenic area's revenue. If SR=1, then the occurrence of public opinion will be marked as having no impact on the local scenic area's revenue; otherwise, SR=1. and set =0;

[0073] The duration of the end of several public opinion events is marked as follows: ;

[0074] Through formula The range influence factor was calculated.

[0075] Through formula The duration of the query interval is calculated.

[0076] S130: Set the format of the j-th token to A+ Within the query time range, sequentially search the public opinion information database for all historical public opinion information corresponding to markers j=n to j=1, obtain the historical public opinion information corresponding to different markers, count the number of historical public opinion information corresponding to different markers, and mark them as... .

[0077] For example, given a scenic area named A, the name of the scenic area is used as the first keyword; custom query terms include: service level, attitude of scenic area service personnel, and environmental sanitation of the scenic area; these custom query terms are used as the second keyword; a public opinion information database is constructed by collecting public opinion information about scenic area A from multiple media such as news, social media, mainstream portals, forums, microblogs, blogs, WeChat official accounts, and OTA platforms; and 20 pieces of public opinion information are selected from the database based on the keywords.

[0078] Public opinion information about the scenic area was retrieved from the public opinion information database. Statistics show that the scenic area experienced negative public opinion twice in the past two years. The incidents occurred on May 13, 2023, lasting for 9 days, and on June 20, 2023, lasting for 5 days. The average duration of the two incidents was 7 days.

[0079] The average revenue of Scenic Area A in June of 2020, 2021, and 2022 was 310,000 yuan, and 310,000 yuan was used as the set low income threshold for June.

[0080] The revenue of Scenic Area A during the most recent public opinion incident in June 2023 was 280,000 yuan, which was lower than the set low revenue threshold of 310,000 yuan for June. Therefore, the public opinion incident in June had an impact on the revenue of Scenic Area A, and this revenue impact was assigned a value of 1.

[0081] Through formula = The calculated range impact factor is 1.7;

[0082] Through formula =7 × 1.7 = The calculated query interval duration is 11.9 days.

[0083] The first tag format is A, and 20 related public opinion messages were found. The second tag format is A+ service level, A+ attitude of scenic area staff, and A+ environmental sanitation of scenic area, and 10 related public opinion messages were found. The third tag format is A+ service level + attitude of scenic area staff and A+ service level + environmental sanitation of scenic area, and 5 related public opinion messages were found. The fourth tag format is A+ service level + attitude of scenic area staff + environmental sanitation of scenic area, and 2 related public opinion messages were found.

[0084] This invention calculates the query interval duration by considering the impact of the most recent public opinion event on the current scenic area; this facilitates more comprehensive and reasonable monitoring of public opinion, enabling tourists to have a more comprehensive understanding of the scenic area and providing them with objective evaluations of public opinion regarding the scenic area.

[0085] Please see Figure 3 The specific steps for training a public opinion type judgment model include:

[0086] S210: Retrieve historical public opinion information corresponding to different markers, and obtain the types of historical public opinion information;

[0087] S220: Retrieve the public opinion type judgment model and mark the number of historical public opinion information corresponding to the marker j=1. ;Will The corresponding historical public opinion information is sequentially input into the public opinion type judgment model; it is judged whether the type of the historical public opinion information output by the public opinion type judgment model is consistent with the type of the marked historical public opinion information; if yes, the output result of the public opinion type judgment model is marked as accurate; if not, the output result of the public opinion type judgment model is marked as inaccurate; the number of public opinion information with accurate output results is counted and recorded as . Among them, the public opinion type judgment model is built based on an artificial intelligence model;

[0088] Through formula The accuracy rate of historical public opinion information when j=1 was calculated.

[0089] S230: Will , , , ..., Each corresponding historical public opinion information is sequentially input into the public opinion type judgment model. Step S220 is repeated to obtain the accuracy of historical public opinion information corresponding to different markers.

[0090] S240: Construct a rectangular coordinate system with the number of keywords as the x-axis and accuracy as the y-axis. Count the total number of curve slopes between adjacent points in the rectangular coordinate system, and the number of curves with a slope greater than or equal to 0 between adjacent points. Determine if the proportion of curves with slopes greater than or equal to 0 to the total number of curves exceeds a threshold. If yes, the training of the public opinion type judgment model is complete; otherwise, use the first item as... The number of historical public opinion information input into the public opinion type judgment model is increased sequentially with a tolerance of 2 until the proportion of the number of curves with a slope greater than or equal to 0 is greater than the proportion threshold, at which point the training process ends.

[0091] For example, retrieve the public opinion type judgment model, and That is, 10 corresponding historical public opinion information are sequentially input into the public opinion type judgment model, and the number of accurate public opinion information output by the public opinion type judgment model is 5.

[0092] Will , , That is, 5, 3, and 1 corresponding historical public opinion information are input into the public opinion type judgment model in sequence, and the number of accurate public opinion information output by the public opinion type judgment model is 3, 2, and 1 respectively;

[0093] Through formula The calculated accuracy rates for public opinion information were 50% when j=1, 60% when j=2, 66.6% when j=3, and 100% when j=5.

[0094] A rectangular coordinate system is constructed with the number of keywords as the x-axis and accuracy as the y-axis. The total number of curve slopes between adjacent points in the rectangular coordinate system is 4; the number of curve slopes between adjacent points in the rectangular coordinate system that are greater than or equal to 0 is 4, and the proportion of curve slopes greater than or equal to 0 to the total number is 100%. Based on the accuracy requirements of professionals in this field for model training, the proportion threshold is set to 95%. Since 100% is greater than 95%, the training of the public opinion type judgment model is completed.

[0095] Please see Figure 4The specific steps for revising public opinion information types include: retrieving public opinion information within the query time range and obtaining a display word cloud of the public opinion information within the query time range; analyzing the sentiment type of several displayed words in the display word cloud based on a sentiment dictionary; where sentiment types include: favorability score of 1, favorability score of 0.5, favorability score of 0, favorability score of -0.5, and favorability score of -1; counting the total number of all displayed words in the display word cloud and the number of displayed words with a favorability score greater than 0 in each sentiment type; and dividing the number of displayed words with a favorability score greater than 0 in each sentiment type by the total number of displayed words to obtain the favorability ratio.

[0096] Determine if the favorability rating is greater than 80%; if yes, mark the corresponding public opinion information as positive; if no, determine if the favorability rating is less than 20%; if yes, mark the corresponding public opinion information as negative.

[0097] No, then mark the corresponding public opinion information as neutral; count the number of likes (DZL), reposts (ZFL), and comments for neutral public opinion information with a favorability ratio within the range of [20%, 40%] ∪ [60%, 80%]; remove invalid comments from the comments; mark the number of comments after removing invalid comments as PLZ, where invalid comments are: comments with fewer than three characters, duplicate comments, and comments whose content is entirely non-text; mark the number of invalid comments as WXL; ∪ represents the union symbol;

[0098] Through formula The public opinion correction factor was calculated; among which, and All are correction factors;

[0099] If the public opinion correction factor is greater than 0, replace the marked neutral public opinion information with a positive one; otherwise, if the public opinion correction factor is less than 0, replace the marked neutral public opinion information with a negative one; otherwise, do not replace the marked neutral public opinion information.

[0100] For example, retrieve public opinion information within the query time range and obtain the displayed word cloud of public opinion information within the query time range, which contains a total of 25 displayed words. Based on the sentiment dictionary analysis, the sentiment type of the displayed words in the displayed word cloud is as follows: 7 words with a favorability score of 1, 8 words with a favorability score of 0.5, 3 words with a favorability score of 0, 4 words with a favorability score of -0.5, and 3 words with a favorability score of -1.

[0101] Statistics show that there are 15 words in the word cloud and 15 words with a positive sentiment score greater than 0 in the sentiment type. Therefore, the positive sentiment score ratio is 15 / 25 = 60%, and the corresponding public opinion information is marked as neutral.

[0102] We retrieved neutral public opinion information whose favorability ratio fell within the range of [20%, 40%] ∪ [60%, 80%]. The number of likes, reposts, and comments corresponding to the neutral public opinion information was 3658, 1279, and 129 respectively. Invalid comments were removed. The number of comments after removing invalid comments was 2541. The number of invalid comments was marked as 129.

[0103] Through formula The calculated public opinion correction factor is 2.7 = 3658. +1279 + Since 2.7 is greater than 0, the public opinion information type marked as neutral will be replaced with positive.

[0104] Please see Figure 5 The second aspect of the present invention provides a smart cultural tourism big data public opinion monitoring and analysis method, comprising:

[0105] A public opinion information database is constructed by collecting public opinion information from multiple media outlets; keywords are set and several pieces of public opinion information are selected from the database based on the keywords; the several pieces of public opinion information are sorted in chronological order of their occurrence; wherein, the keywords include: scenic spot name and custom query terms; the public opinion information includes: the occurrence time of the public opinion, the end time, the scenic spot name, and the custom query terms; and the occurrence time and end time are determined based on the popularity and click-through rate or search rate of the platform where the public opinion is located;

[0106] Construct and train a public opinion type judgment model; determine the type of public opinion information within the query time range based on the public opinion type judgment model; and,

[0107] A public opinion correction factor is constructed to correct the type of public opinion information; the types of public opinion information include: positive, negative and neutral.

[0108] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0109] The working principle of this invention is as follows: collect public opinion information from multiple media to construct a public opinion information database; set keywords and filter out several pieces of public opinion information from the database based on the keywords; sort the several pieces of public opinion information in chronological order of occurrence; construct and train a public opinion type judgment model; determine the type of public opinion information within the query time range based on the public opinion type judgment model; and construct a public opinion correction factor to correct the type of public opinion information.

[0110] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A smart travel big data public opinion monitoring and analyzing system, characterized in that, Comprise: Public opinion collection module and public opinion analysis module connected therewith; Public opinion collection module: collect public opinion information from multiple communication media to construct a public opinion information database; Set keywords and filter a number of public opinion information from the public opinion information database according to the keywords; a number of the public opinion information is sorted according to the chronological order of occurrence; wherein, the keywords include: scenic spot name, custom query words; Public opinion analysis module: construct and train a public opinion type judgment model; determine the type of public opinion information within the query time range according to the public opinion type judgment model; and, Construct a public opinion correction factor to correct the type of public opinion information; wherein, the type of public opinion information includes: positive, negative and neutral; The setting keywords and filtering a number of public opinion information from the public opinion information database, comprising: S110: mark the scenic spot name as the first keyword A, and mark the custom query words as the second keyword Bj; wherein, Bj represents the jth second keyword; j=1, 2, 3, …, m; m is a positive integer; S120: set the query interval duration of public opinion information; S130: set the jth marker format as A+ ; search all historical public opinion information in the query time range from the public opinion information library in sequence, obtain historical public opinion information corresponding to different markers, and count the number of historical public opinion information corresponding to different markers ; The setting query interval duration of public opinion information, comprising: retrieve the public opinion information of the scenic spot from the public opinion information library, count the number k of negative public opinions corresponding to the scenic spot within two years from the current time, and mark the average duration of each historical public opinion as ; obtain the income of the scenic spot when no public opinion occurs every month; wherein, k=0, 1, 2, …, n; n∈[0, +∞]. determining whether the income of the scenic spot in the month when the public opinion information occurs is less than a preset low income threshold; if yes, marking the occurrence of the public opinion as having an impact on the income of the local scenic spot , and setting SR=1; if no, marking the occurrence of the public opinion as having no impact on the income of the local scenic spot , and setting =0; The time length mark of the end of the public opinion of a plurality of pieces of public opinion information is marked as ; The range impact factor is calculated by the formula = 0.5 + 0.5 * (1 - exp(-0.5 The query interval duration is calculated by the formula . 2.The smart travel big data public opinion monitoring and analyzing system according to claim 1, characterized in that, Also include: Public opinion feedback module connected with public opinion analysis module: used to push the type of public opinion information to the user, and used to update the type of public opinion information. 3.The smart travel big data public opinion monitoring and analyzing system according to claim 1, characterized in that, The construction and training of public opinion type judgment model, comprising: S210: call the historical public opinion information corresponding to different markers, and obtain the type of historical public opinion information; S220: call the public opinion type judgment model, mark the number of historical public opinion information corresponding to j=1 as ; input the corresponding historical public opinion information into the public opinion type judgment model in turn; judge whether the type of the historical public opinion information output by the public opinion type judgment model is consistent with the type of the marked historical public opinion information; yes, mark the output result of the public opinion type judgment model as accurate; otherwise, mark the output result of the public opinion type judgment model as inaccurate; count the number of accurate public opinion information and mark it as ; wherein the public opinion type judgment model is constructed based on an artificial intelligence model; The accuracy rate of historical public opinion information corresponding to j=1 is calculated by the formula ​ S230: Will , , , ..., Each corresponding historical public opinion information is sequentially input into the public opinion type judgment model. Step S220 is repeated to obtain the accuracy of historical public opinion information corresponding to different markers. S240: Construct a rectangular coordinate system with the number of keywords as the abscissa and the accuracy rate as the ordinate, count the total number of curve slopes between adjacent two points in the rectangular coordinate system, and the number of curve slopes between adjacent two points in the rectangular coordinate system greater than or equal to 0; determine whether the proportion of the number of curve slopes greater than or equal to 0 to the total number is greater than the proportion threshold; yes, complete the training of the public opinion type judgment model; no, take the first item as , the tolerance is 2, the number of historical public opinion information input into the public opinion type judgment model is increased in turn until the proportion of the number of curve slopes greater than or equal to 0 to the total number is greater than the proportion threshold, and the process of continuing training is ended. 4.The smart travel big data public opinion monitoring and analyzing system according to claim 3, characterized in that, The public opinion type judgment model is constructed based on artificial intelligence model, comprising: Obtain standard training data; integrate all historical public opinion information into a display word cloud map through word cloud map generation technology, and arrange and integrate the display words in the display word cloud map into a judgment factor sequence in descending order of font size; wherein, the standard training data includes standard input data consistent with the content attribute of the judgment factor sequence, and standard output data consistent with the content attribute of the judgment factor sequence; the standard output data is the type of public opinion information; train the artificial intelligence model using the standard training data; Mark the trained artificial intelligence model as the public opinion type judgment model; wherein, the artificial intelligence model includes convolutional neural network model or long short memory neural network model. 5.The smart travel and leisure big data public opinion monitoring and analyzing system according to claim 1, characterized in that, The determination of the type of public opinion information within the query time range according to the public opinion type judgment model, comprising: Call the public opinion information within the query time range, obtain the display word cloud map of the public opinion information within the query time range, and analyze the emotional type of a number of display words in the display word cloud map based on the sentiment dictionary; wherein, the emotional type includes: good feeling a, good feeling b, good feeling 0, good feeling-b, good feeling-a; count the number of all display words in the display word cloud map and the number of display words with good feeling greater than 0 in the emotional type; divide the number of display words with good feeling greater than 0 in the emotional type by the number of all display words to obtain the good feeling ratio; wherein, a>b>0; Divide the type of public opinion information according to the comparison threshold. 6.The smart travel and leisure big data public opinion monitoring and analyzing system according to claim 1, characterized in that, The construction of public opinion correction factor to correct the type of public opinion information, comprising: The like quantity DZL, the forwarding quantity ZFL and the comment quantity of the public opinion information corresponding to the public opinion information type of neutral are selected, the invalid comments in the comment quantity are removed, the comment quantity after removing the invalid comments is marked as PLZ, and the number of invalid comments is marked as WXL; The public opinion correction factor is calculated by the formula wherein, and are correction coefficients. It is judged whether the public opinion correction factor is greater than 0; if yes, the public opinion information marked as neutral is replaced by the positive; if no, it is judged whether the public opinion correction factor is less than 0; if yes, the public opinion information marked as neutral is replaced by the negative; if no, the public opinion information marked as neutral is not replaced. 7.The smart travel big data public opinion monitoring and analyzing system of claim 6, wherein, The correction coefficient is obtained in the following manner: The bit number e of the like quantity DZL value and the bit number w of the forwarding quantity ZFL value are obtained, The correction factor α is calculated as follows: α = ; The calculation method of the correction coefficient β is: β = e / w ; wherein, the value range of e and w is [0, +∞].

8. A smart travel big data public opinion monitoring and analysis method, suitable for a smart travel big data public opinion monitoring and analysis system according to any one of claims 1-7, characterized in that, It comprises: Collecting public opinion information from multiple propagation media to construct a public opinion information library; Setting keywords and screening a plurality of pieces of public opinion information from the public opinion information library according to the keywords; The plurality of pieces of public opinion information are sorted according to the chronological order; wherein the keywords include: scenic spot name, self-defined query word; A public opinion type judgment model is constructed and trained, the type of the public opinion information in the query time range is determined according to the public opinion type judgment model; and A public opinion correction factor is constructed to correct the type of the public opinion information; wherein the type of the public opinion information includes: positive, negative and neutral.

Images