An online big data-based comprehensive transportation hub service perception intelligent monitoring method and system

Abstract

The application relates to the technical field of intelligent traffic management, and discloses a comprehensive traffic hub service perception intelligent monitoring method and system based on online big data, which comprises the following steps: obtaining an effective social media text sample set; performing service perception theme identification based on the effective social media text sample set; performing sentiment tendency division on the effective social media text; constructing a service attribute dictionary; performing service attribute labeling and service bottleneck identification based on the service attribute dictionary; performing time sequence dynamic analysis of service perception based on a sentiment probability value and a publishing time of a user on a social media platform to obtain a service deficiency period of the comprehensive traffic hub; and outputting the identified service bottleneck attribute and the obtained service deficiency period. The application can continuously and low-costly obtain real perception feedback of passengers on traffic service of a comprehensive passenger transport hub without relying on traditional questionnaire investigation, and provides data support for hub operation management, service optimization and emergency response.

Description

Technical Field

[0001] This invention belongs to the field of intelligent traffic management technology, and relates to an intelligent monitoring method and system for comprehensive transportation hub service perception based on online big data, which is used to characterize the subjective perception characteristics of transportation hub users to travel services and their temporal change patterns. Background Technology

[0002] As crucial nodes in urban and regional transportation systems, integrated transportation hubs serve as hubs for the distribution of passengers across modes of transport and regions. Their service quality directly impacts passenger travel experience and the operational efficiency of the transportation system. With the continuous expansion of the scale and functions of integrated transportation hubs, passengers' focus on hub services has gradually shifted from solely addressing accessibility to a comprehensive perception encompassing multiple dimensions, including transfer convenience, information accessibility, and service reliability.

[0003] Existing methods for assessing the service quality of transportation hubs mainly rely on traditional approaches such as questionnaires and on-site interviews. While these methods have advantages such as clear structure and controllable indicators, they generally suffer from limitations such as limited sample size, long survey periods, and difficulty in reflecting real-time changes, making it difficult to meet the actual needs of complex transportation hubs for continuous and dynamic service monitoring.

[0004] In recent years, social media platforms have become an important channel for travelers to express their travel experiences and service evaluations. The text information spontaneously posted by users on these platforms is characterized by its large sample size, continuous temporal coverage, and direct emotional expression, providing a new data source for research on service perception in transportation hubs. However, current applications of social media data mostly remain at the level of sentiment polarity statistics or simple topic identification, lacking a systematic analysis of the service perception structure, key service attributes, and their temporal evolution characteristics. This makes it difficult to directly provide actionable decision support for the operation and management of transportation hubs. Summary of the Invention

[0005] To address the problems of existing integrated passenger transport hub service quality assessments relying on manual surveys, lacking timeliness, and failing to reflect the dynamic changes in passengers' actual perceptions, this invention provides an intelligent monitoring method and system for integrated transportation hub service perception based on online big data. This method can continuously and cost-effectively obtain passengers' real perception feedback on integrated passenger transport hub transportation services without relying on traditional questionnaires, providing data support for hub operation management, service optimization, and emergency response, and improving the precision of integrated passenger transport hub operation management and decision-making response capabilities.

[0006] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution.

[0007] Firstly, this invention proposes a comprehensive intelligent monitoring method for transportation hub services based on online big data, comprising:

[0008] Step S1: Collect social media text data to obtain the original social media text dataset.

[0009] Taking integrated transportation hubs as the research object, keywords or keyword combinations that uniquely correspond to the integrated transportation hubs are identified; based on the keywords, web crawling technology is used to obtain text data published by users from social media platforms; the text data includes at least the text content published by the user, the publication time, and user identification information, and the text data is continuously collected according to a preset time range (i.e., a set time span) to construct the original social media text dataset.

[0010] Step S2 involves cleaning and filtering the obtained raw social media text dataset to obtain a valid social media text sample set.

[0011] The following preprocessing operations are performed on the original social media text dataset:

[0012] (1) Remove texts published by institutional accounts, marketing accounts and non-personal user accounts, and remove texts from accounts that post frequently (high frequency of posting refers to an average monthly posting volume of more than 40) but whose content is mainly work or business.

[0013] (2) Remove text content that is irrelevant to actual travel or use of integrated transportation hubs;

[0014] (3) For the same text repeatedly published by the same user, remove the repeated identical text and only retain the earliest publication time record;

[0015] (4) Based on the preliminary topic analysis results (i.e., based on the topic modeling results of the pre-experiment), remove texts with semantic anomalies or that are irrelevant to travel;

[0016] Through the above processing, a valid social media text sample set is obtained for subsequent analysis.

[0017] Step S3: Based on the effective social media text sample set, perform service-aware topic identification to obtain several service-aware topics.

[0018] The effective social media text samples are subjected to text preprocessing such as word segmentation, stop word filtering, punctuation and noise removal to obtain preprocessed text;

[0019] An unsupervised topic modeling method is used to model and analyze the preprocessed text to uncover potential service-aware topics that users are interested in.

[0020] Based on the topic consistency index and the results of manual interpretation, determine the optimal number of topics and output several service-aware topics;

[0021] Several service perception themes are used as different service dimensions of integrated transportation hub service perception to represent the service content that users care about.

[0022] Step S4: Based on the effective social media text sample set, obtain the sentiment probability values ​​of service perception topics related to traffic operation and travel process, and then classify the effective social media texts into sentiment tendencies.

[0023] Valid social media text samples are input into a sentiment analysis tool, which outputs a sentiment probability value for each valid social media text. The sentiment probability value is used to characterize the user's sentiment tendency towards the corresponding service's perceived theme (dimension); based on preset thresholds, valid social media texts are divided into three sentiment tendencies: positive sentiment (≥0.55), neutral sentiment (0.45-0.55), and negative sentiment (≤0.45).

[0024] Step S5: Construct a service attribute dictionary for integrated transportation hubs, targeting service perception themes related to traffic operation and travel process; perform service attribute annotation and service bottleneck identification based on the service attribute dictionary to obtain annotated text and identified potential service bottleneck attributes.

[0025] A service attribute dictionary is constructed for transportation hub services. The service attribute dictionary includes at least the following attributes: fare-related attributes, accessibility and connection attributes, availability attributes, congestion level attributes, transfer convenience attributes, security check process attributes, and baggage check-in and retrieval attributes. Service attributes are labeled for valid social media texts, allowing a single valid social media text to be associated with multiple service attributes. The distribution of sentiment tendencies of texts under different service attributes is statistically analyzed, that is, the proportion of valid social media texts with negative sentiment for all service attributes is statistically analyzed, and the service attributes with the top 20% of negative sentiment proportions among all service attributes are regarded as service bottleneck attributes in the integrated transportation hub.

[0026] Step S6: Based on service satisfaction (emotional probability value) and the posting time on users' social media platforms, conduct a time-series dynamic analysis of service perception to obtain the periods of insufficient service at the integrated transportation hub.

[0027] Based on a preset time granularity: by hourly segments (divided into 6-hour time groups: late night: 0-5:59 AM, morning: 6:00-11:59 AM, afternoon: 12:00-5:59 PM, evening: 6:00-11:59 PM) and a seven-day week (Monday to Sunday), the average user satisfaction for each time period is calculated. Based on this average service satisfaction for each time period, a service satisfaction deviation index is constructed to measure the degree of change in service perception (satisfaction) relative to the overall average level within the preset time granularity; that is, the degree of deviation of the emotional score for each time period group from the average of that time period group over the seven days of the week. This service perception deviation index is defined as the "satisfaction deviation index". The purpose is to provide quantifiable decision-making basis for relevant policy formulation, as shown in the following formula.

[0028] ;

[0029] in, This indicates the number of positive tweets (positive tweets refer to valid social media texts categorized as having positive sentiment) within a specific time period (e.g., 12:00–17:59) on a particular date (e.g., Monday); This represents the total number of positive tweets over a seven-day week during that period. Similarly, This indicates the number of negative tweets (negative tweets refer to valid social media texts categorized as having negative sentiment) within a specific time period (e.g., 12:00–17:59) on a given date (e.g., Monday). This represents the total number of negative tweets within a seven-day period for that time group (if there are no negative tweets in a certain time period, the divisor is used). When it is 0, then Set to 0.1 to avoid calculation errors.

[0030] The higher the value, the better the service perception evaluation during that period; The smaller the value, the worse the service perception evaluation during that period.

[0031] When the emotional tendency is positive, if Greater than or equal to the preset threshold When the time is right, it is a period of service surplus; conversely, when the time is wrong, it is a period of service deficit. Preferably, That is, when A score greater than 0.3 indicates that the emotional score for that period deviates by more than 30% compared to the average level for the same period within a week. This invention will... A value greater than 0.3 is defined as a period of service surplus; conversely, a value less than 0.3 is defined as a period of service shortage.

[0032] Similarly, when the emotional inclination is positive, if Greater than or equal to When the time is right, it is a period of surplus service; conversely, when the time is wrong, it is a period of underserved service.

[0033] Step S7: Output the integrated transportation hub service perception monitoring results.

[0034] The service bottleneck attributes identified in step S5 and the results of the time periods of insufficient service obtained in step S6 are output to support service improvement of integrated transportation hubs.

[0035] This invention uses automatically collected text information related to integrated transportation hubs posted by passengers on social media platforms as a data source. Through automated collection and processing, the text data is cleaned and preprocessed to identify user-perceived service perception themes, label specific service attributes under these themes, determine sentiment tendencies, and extract service attributes reflecting service bottlenecks. Based on this, it dynamically characterizes service perception satisfaction (i.e., service perception deviation index) across different time periods, identifying key periods within a week where service satisfaction is significantly lower than other similar periods, enabling refined monitoring and early warning of service fluctuations. This invention can continuously and cost-effectively obtain passengers' genuine perception feedback on integrated transportation hub services without relying on traditional questionnaires, providing data support for hub operation management, service optimization, and emergency response, thereby improving the refinement level and decision-making response capabilities of integrated transportation hub operation management.

[0036] Secondly, this invention proposes an intelligent monitoring system for integrated transportation hub services based on online big data, used to implement the aforementioned intelligent monitoring method for integrated transportation hub services based on online big data, comprising:

[0037] The data acquisition module is configured to collect social media text data to obtain the raw social media text dataset;

[0038] The text cleaning and sample filtering module is configured to clean and filter the obtained raw social media text dataset to obtain a valid social media text sample set;

[0039] The service-aware topic extraction module is configured to identify service-aware topics based on a valid social media text sample set, and obtain several service-aware topics.

[0040] The text sentiment analysis module is configured to obtain sentiment probability values ​​for service perception topics related to traffic operation and travel process based on a valid social media text sample set, and then classify the sentiment tendency of valid social media texts.

[0041] The service bottleneck identification module is configured to construct a service attribute dictionary for integrated transportation hubs, targeting service perception themes related to traffic operation and travel process; and to perform service attribute annotation and service bottleneck identification based on the service attribute dictionary, thereby obtaining the annotated text and the identified service bottleneck attributes.

[0042] The service shortage period identification module is configured to perform time-series dynamic analysis of service perception based on sentiment probability values ​​and the posting time on users' social media platforms to obtain the service shortage periods of integrated transportation hubs.

[0043] The output module is configured to output the identified service bottleneck attributes and the results of the service insufficiency period, that is, to output the service perception monitoring results of the integrated transportation hub.

[0044] Thirdly, the present invention proposes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described intelligent monitoring method for integrated transportation hub services based on online big data.

[0045] Fourthly, the present invention provides a computer device comprising:

[0046] Memory, used to store computer programs;

[0047] A processor is used to execute the computer program to implement the steps of the above-described intelligent monitoring method for integrated transportation hub services based on online big data.

[0048] Fifthly, the present invention proposes a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described intelligent monitoring method for integrated transportation hub services based on online big data.

[0049] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0050] (1) Based on social media text data, this invention realizes continuous and dynamic monitoring of service perception of integrated transportation hubs, overcoming the shortcomings of traditional questionnaire surveys with small sample size and long update cycle; and this invention does not require additional sensor deployment or large-scale surveys, and has the advantages of low implementation cost and wide applicability, and is suitable for service perception monitoring scenarios of various types of integrated transportation hubs.

[0051] (2) By combining topic modeling and sentiment analysis, this invention can automatically identify the service perception dimensions that users care about and quantify the impact of different service attributes on the overall perception, thereby improving the level of service evaluation.

[0052] (3) The present invention introduces a time-series dynamic analysis method, which can identify the fluctuation characteristics of service perception in different time periods, and provide a scientific basis for time-series service management and long-term service optimization of integrated transportation hubs.

[0053] (4) The present invention can make full use of social media text data to automatically identify the service perception dimension of transportation hubs, quantify user emotional tendencies, and reveal the law of service perception change over time, thus making up for the shortcomings of existing technologies in dynamic perception and refined analysis. Attached Figure Description

[0054] Figure 1 This is a flowchart illustrating the monitoring method in Embodiment 1 of the present invention. Detailed Implementation

[0055] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0056] The term "and / or" simply describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0057] Example 1

[0058] This embodiment takes a transportation hub as the research object and implements a comprehensive transportation hub service perception dynamic monitoring method based on social media big data, based on travel-related text information posted by users on social media platforms. This method is used to automatically identify the service perception dimensions that users pay attention to, quantify sentiment tendencies, and characterize their changes over time, thereby providing a basis for decision-making in transportation hub operation management and service optimization.

[0059] like Figure 1 As shown, the steps of the integrated transportation hub service perception and intelligent monitoring method based on online big data in this embodiment are as follows:

[0060] Step S1: Obtain social media text data. Collect social media text data to obtain the raw social media text dataset.

[0061] In this embodiment, "a certain airport" was selected as the core search term. Text data posted by users was continuously collected from social media platforms using web crawlers, with the collection time span covering multiple calendar years to ensure the representativeness of the samples across different time periods and operating states. The collected data included at least: the text content posted by users, the corresponding posting time information, and user identification information; thus, an original social media text dataset was constructed. In this embodiment, approximately 30,000 original social media text samples were collected.

[0062] Step S2, Text Cleaning and Sample Filtering. The obtained raw social media text dataset is cleaned and filtered to obtain a valid social media text sample set.

[0063] To address the issues of high noise levels and inconsistent quality in raw social media text datasets, the following cleaning and filtering steps are performed on the raw social media text datasets:

[0064] Step S21: Remove texts from accounts including institutional accounts, marketing accounts, and non-travel individual accounts; remove texts from accounts that post frequently (high frequency of posting refers to an average monthly posting volume of more than 40) but whose content is mainly work or business.

[0065] Step S22: Based on keywords and semantic rules, remove text content that is irrelevant to actual travel or use of integrated transportation hubs;

[0066] Step S23: For the same text repeatedly published by the same user, remove the duplicate text and only retain the earliest publication time record;

[0067] Step S24: Based on the topic modeling results of the pre-experiment, remove text with semantic anomalies or that is irrelevant to travel.

[0068] It should be noted that the topic modeling results of the pre-experiment are obtained by directly using existing open-source algorithms, which can be called to obtain service-aware topic analysis results, i.e., the topic modeling results of the pre-experiment.

[0069] Through the above multi-stage screening process, an effective social media text sample set was obtained for subsequent analysis.

[0070] Step S3: Document preprocessing and service-aware topic extraction. Based on a valid set of social media text samples, service-aware topic identification is performed to obtain several service-aware topics.

[0071] Step S31: Perform text preprocessing operations such as word segmentation, stop word filtering, punctuation and noise removal on the valid social media text samples to obtain the preprocessed text;

[0072] Step S32: The preprocessed text input topic modeling algorithm (i.e., the latent Dirichlet assignment model) is used to mine the latent semantic structure hidden in the user's text to obtain the latent service-aware topics that the user is interested in.

[0073] Step S33: Based on the potential service-aware topics that users are interested in, determine the optimal number of topics through topic consistency index, and combine manual semantic interpretation to summarize and name the determined topics, and output several service-aware topics.

[0074] Among them, the service perception theme, as different service dimensions of integrated transportation hub service perception, is used to characterize the service content that users care about.

[0075] In this embodiment, four types of service perception themes are identified: air-rail integrated travel-related services, airport services and supporting facilities, hub building space and design perception, and hub internal leisure and ancillary activities.

[0076] The first two categories of service perception themes are directly related to traffic operation and the travel process, and are therefore the key areas of focus for hub managers. The latter two categories of service perception themes reflect the service perception characteristics of integrated transportation hubs as public spaces, are unrelated to traffic, and are not used for hub service monitoring.

[0077] Step S4, Text Sentiment Analysis. Based on a valid social media text sample set, the sentiment probability value of the service-perceived topic is obtained, and then the valid social media text is classified into sentiment tendencies.

[0078] Step S41: Input the valid social media text samples obtained in step S2 into the sentiment analysis tool (i.e., the sentiment analysis model pre-trained by the model developer, which is an existing tool and can be directly called), and output the corresponding sentiment probability value for each valid social media text.

[0079] The emotional probability value, or service satisfaction value, is used to characterize the intensity of positive or negative emotions expressed in the text, and its value ranges from 0 to 1.

[0080] Step S42: Based on the sentiment probability value, the effective social media texts are divided into three sentiment tendencies: positive sentiment (0.55≤sentiment probability value≤1), neutral sentiment (0.45<sentiment probability value<0.55), and negative sentiment (0≤sentiment probability value≤0.45).

[0081] Step S5, Service Bottleneck Identification. For the service perception theme, a service attribute dictionary for integrated transportation hubs is constructed. Service attribute annotation and service bottleneck identification are performed to obtain the annotated text and the identified potential service bottleneck attributes.

[0082] Step S51: Construct a service attribute dictionary for integrated transportation hubs, targeting service perception themes directly related to traffic operation and travel process. The service attribute dictionary shall include at least: fare-related attributes, accessibility and connection attributes, availability attributes, transfer convenience attributes, congestion level attributes, security check process attributes, and baggage check-in and retrieval attributes.

[0083] Step S52: Label the valid social media text with service attributes, allowing a single text to be associated with multiple service attributes;

[0084] Step S523: Statistically analyze the sentiment distribution of text under different service attributes, identify attributes with a high proportion of negative sentiment (referring to service attributes with a proportion of negative sentiment ranking in the top 20% of all service attributes), and use them as service bottleneck factors (i.e. service bottleneck attributes).

[0085] Step S6, Identification of Service Shortage Periods. Based on service satisfaction (i.e., sentiment probability value) and the posting time on users' social media platforms, a time-series dynamic analysis of service perception is conducted to identify weak periods where the service satisfaction deviation index of the integrated transportation hub is low (i.e., service is insufficient).

[0086] Step S61: According to the preset time granularity: by hourly segments (divided into 6-hour time groups, namely late night: 0-5:59, morning: 6:00-11:59, afternoon: 12:00-17:59, evening: 18:00-23:59) and the seven-day scale of a week (Monday to Sunday), calculate the average user satisfaction for each time period;

[0087] Step S62: Identify the weak periods in service satisfaction assessment. Based on the average service satisfaction rate for each period, construct a service perception deviation index to measure the degree of change in service satisfaction relative to the overall average level within a preset time granularity. This means that the emotional probability value of each period group deviates from the average value of that period group over seven days of the week. This service perception deviation index is defined as the "satisfaction deviation index". The purpose is to provide quantifiable decision-making basis for relevant policy formulation, as shown in the following formula.

[0088] ;

[0089] in, This indicates the number of positive tweets on a specific date (e.g., Monday) within a specific time period (e.g., 12:00–17:59). This represents the total number of positive tweets over a seven-day week during that period. Similarly, This indicates the number of negative tweets on a specific date (e.g., Monday) within a specific time period (e.g., 12:00–17:59). This represents the total number of negative tweets within a seven-day period for that time group (if there are no negative tweets in a certain time period, the divisor is used). When it is 0, then Set to 0.1 to avoid calculation errors.

[0090] The higher the value, the better the service perception evaluation during that period; The smaller the value, the worse the service perception evaluation during that period.

[0091] When the emotional tendency is positive, if Greater than or equal to the preset threshold When the time is right, it is a period of service surplus; conversely, when the time is wrong, it is a period of service deficit. Preferred That is, when When the emotional score is greater than the preset threshold of 0.3, it indicates that the emotional score of this period deviates from the average level of the same period within a week by more than 30%, and this invention defines it as a service surplus period; otherwise, it is a service undersupplied period.

[0092] Similarly, when the emotional inclination is positive, if Greater than or equal to When the time is right, it is a period of service surplus; conversely, when the time is wrong, it is a period of service deficit. Preferred That is, when When the score is greater than the preset threshold of -0.3, it indicates that the emotional score of that period deviates by more than 30% compared with the average level of the same period within a week. This invention defines it as a period of service surplus; otherwise, it is a period of service shortage.

[0093] The above methods are used to identify periods when service perception evaluations are significantly low or high, in order to support dynamic service monitoring and management decisions for transportation hubs.

[0094] Step S7, output the result.

[0095] The service perception satisfaction and its temporal variation characteristics under the service perception theme (i.e., the service bottleneck attributes identified in step S5 and the time periods of service insufficiency obtained in step S6) together constitute the service perception monitoring results of the integrated transportation hub of the present invention; the obtained integrated transportation hub service perception monitoring results can reflect the main service dimensions that users pay attention to, the differences in emotional tendencies of different service attributes, and the changing characteristics of service perception in the time dimension.

[0096] Through the above embodiments, automated identification and dynamic monitoring of service perception of integrated transportation hubs can be achieved. The monitoring results of integrated transportation hub service perception can be used to identify weak links in the operation of transportation services; support service optimization and resource allocation at different times; and provide quantitative basis for continuous operation monitoring and management decisions of integrated transportation hubs.

[0097] Example 2

[0098] This embodiment further illustrates the intelligent monitoring method for integrated transportation hub services based on online big data using more specific text / data.

[0099] Taking the service perception theme "integrated air-rail travel related services" identified in Example 1 as an example, two valid social media texts are selected for processing and explanation.

[0100] Example text 1:

[0101] Social media text: "First experience on an airport line, so cool (mainly because I suddenly saw that an airline was giving away light rail tickets before entering the station, so I got a 35 yuan discount, and I'm so happy now haha)."

[0102] The steps are explained below:

[0103] Step 1: The service attributes are manually annotated. This text involves "transfer fare related attributes".

[0104] Step 2, Sentiment Analysis Processing: Input the text into the sentiment analysis tool, which outputs a sentiment probability value of 0.999. Based on the sentiment threshold set in Example 1 (≥0.55 for positive sentiment), it is determined to be positive sentiment.

[0105] Step 3, Time Information Extraction: The release time is Saturday morning, so it is classified into the "Morning (6:00–12:00)" time period group and the "Saturday" category.

[0106] Step 4, the text is therefore categorized as:

[0107] Service perception theme: Services related to integrated air-rail travel;

[0108] Service attributes: Ticket price related attributes;

[0109] Emotional tendency: Positive emotion;

[0110] Time tag: Saturday morning.

[0111] Example text 2:

[0112] Social media text: "A subway line at an airport feels more like a high-speed rail. The seats are single, it runs on the ground, the stations are far apart, and the tickets are expensive. Is it a city high-speed rail?"

[0113] The steps are explained below:

[0114] Step 1: Service attributes are manually labeled. This text also includes: fare-related attributes and transfer convenience attributes.

[0115] Step 2, Sentiment Analysis Processing: After inputting into the sentiment analysis tool, the output sentiment probability value is 0.012. Based on the sentiment threshold set in Example 1 (≤0.45 is negative sentiment), it is determined to be negative sentiment.

[0116] Step 3, Time Information Extraction: The release time is Tuesday morning, so it is classified into the "Morning (6:00–12:00)" time period group and the "Tuesday" category.

[0117] Step 4, the text is therefore categorized as:

[0118] Service perception theme: Services related to integrated air-rail travel;

[0119] Service attributes: fare-related attributes, transfer convenience attributes;

[0120] Emotional tendency: Negative emotions;

[0121] Time tag: Tuesday morning.

[0122] Once a large amount of text has undergone the above steps for service-aware topic identification, service attribute labeling, and sentiment quantification, statistical analysis can be performed on the sentiment distribution under different service attributes. For example, assuming there are 10 service attributes, the top 20% of service attributes correspond to the top 2 (calculated as...). Assuming the negative sentiment percentage for transfer convenience is 42%, the negative sentiment percentage for fare-related attributes is 56%, and the negative sentiment percentages for other service attributes are 25%, 30%, etc., respectively. If the negative sentiment percentage for fare-related attributes is higher than that for other service attributes, and ranks second in the negative sentiment percentage ranking (i.e., entering the top 20%), then according to the rule in step S53 of Example 1, "fare-related attributes" are identified as potential service bottleneck attributes. Simultaneously, in the time dimension: if the satisfaction index under the service perception theme on "Tuesday morning" deviates from the index (… If the value is lower than a preset threshold (e.g., less than -0.3), then the period is determined to be a period of insufficient service.

[0123] Example 3

[0124] Based on the same inventive concept as Embodiment 1, this embodiment introduces a comprehensive transportation hub service perception and intelligent monitoring system based on online big data, including:

[0125] The data acquisition module is configured to collect social media text data to obtain the raw social media text dataset;

[0126] The text cleaning and sample filtering module is configured to clean and filter the obtained raw social media text dataset to obtain a valid social media text sample set;

[0127] The service-aware topic extraction module is configured to identify service-aware topics based on a valid social media text sample set, and obtain several service-aware topics.

[0128] The text sentiment analysis module is configured to obtain sentiment probability values ​​for service perception topics related to traffic operation and travel process based on a valid social media text sample set, and then classify the sentiment tendency of valid social media texts.

[0129] The service bottleneck identification module is configured to construct a service attribute dictionary for integrated transportation hubs, targeting service perception themes related to traffic operation and travel process; and to perform service attribute annotation and service bottleneck identification based on the service attribute dictionary, thereby obtaining the annotated text and the identified service bottleneck attributes.

[0130] The service shortage period identification module is configured to perform time-series dynamic analysis of service perception based on sentiment probability values ​​and the posting time on users' social media platforms to obtain the service shortage periods of integrated transportation hubs.

[0131] The output module is configured to output the identified service bottleneck attributes and the results of the service insufficiency period, that is, to output the service perception monitoring results of the integrated transportation hub.

[0132] Example 4

[0133] Based on the same inventive concept as other embodiments, this embodiment introduces a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described integrated transportation hub service perception and intelligent monitoring method based on online big data.

[0134] Example 5

[0135] Based on the same inventive concept as other embodiments, this embodiment introduces a computer device, including: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the above-described integrated transportation hub service perception and intelligent monitoring method based on online big data.

[0136] Example 6

[0137] Based on the same inventive concept as other embodiments, this embodiment introduces a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described integrated transportation hub service perception and intelligent monitoring method based on online big data.

[0138] In summary, this invention enables automated and dynamic monitoring of service perception at transportation hubs by performing service perception topic identification, sentiment analysis, and temporal modeling on user-published text information.

[0139] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0140] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0141] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0142] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0143] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other modifications under the guidance of the present invention, and these modifications are all within the protection scope of the present invention.

Claims

1. A comprehensive transportation hub service perception and intelligent monitoring method based on online big data, characterized in that, include: Collect social media text data to obtain the raw social media text dataset; The obtained raw social media text dataset is cleaned and filtered to obtain a valid social media text sample set; Based on a valid social media text sample set, service perception theme identification was performed to obtain several service perception themes; Based on a sample set of effective social media texts, we obtain the sentiment probability values ​​of service perception topics related to traffic operation and travel process, and then classify the sentiment tendencies of effective social media texts. For service perception topics related to traffic operation and travel process, a service attribute dictionary for integrated transportation hubs is constructed; based on the service attribute dictionary, service attribute annotation and service bottleneck identification are performed to obtain the annotated text and the identified service bottleneck attributes. Based on sentiment probability values ​​and the posting time on users' social media platforms, a time-series dynamic analysis of service perception is conducted to identify periods of service shortage at integrated transportation hubs. The identified service bottleneck attributes and the results of service insufficiency periods are output, which is the output of the integrated transportation hub service perception monitoring results.

2. The intelligent monitoring method for integrated transportation hub services based on online big data according to claim 1, characterized in that, The process of cleaning and filtering the obtained raw social media text dataset yields a valid social media text sample set, including: Texts published by institutional accounts, marketing accounts, and non-personal user accounts will be removed. Remove text that is irrelevant to actual travel or use at integrated transportation hubs; For the same text repeatedly posted by the same user, the duplicate posts are removed, and only the earliest posting time is retained. Based on the thematic analysis results, texts with semantic anomalies or irrelevant to travel are removed; Through the above processing, a valid social media text sample set is obtained.

3. The intelligent monitoring method for integrated transportation hub services based on online big data according to claim 1, characterized in that, The process of identifying service-aware topics based on a valid social media text sample set yields several service-aware topics, including: The effective social media text is preprocessed to obtain the preprocessed text. An unsupervised topic modeling method is used to model and analyze the preprocessed text to obtain potential service-aware topics that users are interested in; Based on potential service awareness topics that have attracted user attention, the optimal number of topics is determined according to topic consistency indicators and manual interpretation results, and several service awareness topics are output.

4. The intelligent monitoring method for integrated transportation hub services based on online big data according to claim 1, characterized in that, Based on a valid social media text sample set, sentiment probability values ​​for service perception topics related to traffic operation and travel process are obtained, and then the valid social media texts are classified into sentiment tendencies, including: Valid social media texts are input into a sentiment analysis tool, which outputs a sentiment probability value for each valid social media text; the sentiment probability value is used to characterize a user's sentiment tendency towards a certain service's perceived theme. Based on sentiment probability values ​​and set sentiment thresholds, effective social media texts are categorized into positive sentiment, neutral sentiment, and negative sentiment. The method for setting the emotion threshold is as follows: an emotion probability value ≥ 0.55 indicates a positive emotion; an emotion probability value of 0.45-0.55 indicates a neutral emotion; and an emotion probability value ≤ 0.45 indicates a negative emotion.

5. The intelligent monitoring method for integrated transportation hub services based on online big data according to claim 1, characterized in that, The process involves constructing a service attribute dictionary for integrated transportation hubs, targeting service perception themes related to traffic operation and travel processes. Based on this dictionary, service attribute annotation and service bottleneck identification are performed to obtain annotated text and identified service bottleneck attributes, including: The service attribute dictionary includes fare-related attributes, accessibility and connection attributes, availability attributes, congestion level attributes, transfer convenience attributes, security check process attributes, and baggage check-in and retrieval attributes; service attributes are labeled for valid social media texts, allowing each valid social media text to be associated with at least one service attribute; the distribution of sentiment tendencies of texts under different service attributes is statistically analyzed, and the top 20% of service attributes with negative sentiment are used as service bottleneck attributes in integrated transportation hubs.

6. The intelligent monitoring method for integrated transportation hub services based on online big data according to claim 1, characterized in that, The time-series dynamic analysis of service perception based on sentiment probability values ​​and the posting time on users' social media platforms is used to obtain the periods of service insufficiency at integrated transportation hubs, including: The preset time granularity is used to calculate the average user satisfaction for each time period, based on hourly segments and seven days within a week. Based on the average service satisfaction rate over different time periods, a service satisfaction deviation index is constructed. The specific expression is: ； in, This indicates the number of positive tweets within a specific time period on a given date. This represents the total number of positive tweets within a seven-day week during that period; similarly, This indicates the number of negative tweets within a specific time period on a given date. This represents the total number of negative tweets in that time period over seven days. The higher the value, the better the service perception evaluation during that period; The smaller the value, the worse the service perception evaluation during that period; When the emotional tendency is positive, if Greater than or equal to When the time is right, it is a period of service surplus; conversely, when the time is wrong, it is a period of service deficit. When the emotional tendency is positive, if Greater than or equal to When the time is right, it is a period of surplus service; conversely, when the time is wrong, it is a period of underserved service.

7. A comprehensive transportation hub service perception and intelligent monitoring system based on online big data, characterized in that, The method for implementing the above-mentioned intelligent monitoring of integrated transportation hub services based on online big data includes: The data acquisition module is configured to collect social media text data to obtain the raw social media text dataset; The text cleaning and sample filtering module is configured to clean and filter the obtained raw social media text dataset to obtain a valid social media text sample set; The service-aware topic extraction module is configured to identify service-aware topics based on a valid social media text sample set, and obtain several service-aware topics. The text sentiment analysis module is configured to obtain sentiment probability values ​​for service perception topics related to traffic operation and travel process based on a valid social media text sample set, and then classify the sentiment tendency of valid social media texts. The service bottleneck identification module is configured to construct a service attribute dictionary for integrated transportation hubs, targeting service perception themes related to traffic operation and travel process; and to perform service attribute annotation and service bottleneck identification based on the service attribute dictionary, thereby obtaining the annotated text and the identified service bottleneck attributes. The service shortage period identification module is configured to perform time-series dynamic analysis of service perception based on sentiment probability values ​​and the posting time on users' social media platforms to obtain the service shortage periods of integrated transportation hubs. The output module is configured to output the identified service bottleneck attributes and the results of the service insufficiency period, that is, to output the service perception monitoring results of the integrated transportation hub.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the integrated transportation hub service perception and intelligent monitoring method based on online big data as described in any one of claims 1 to 6.

9. A computer device, characterized in that, include: Memory, used to store computer programs; A processor is configured to execute the computer program to implement the steps of the integrated transportation hub service perception and intelligent monitoring method based on online big data as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that: When the computer program is executed by the processor, it implements the steps of the integrated transportation hub service perception and intelligent monitoring method based on online big data as described in any one of claims 1 to 6.

Images