Method and device of online car-hailing driver and passenger mutual evaluation system based on deep learning

Abstract

The invention relates to a method and a device of an online car-hailing driver and passenger mutual evaluation system based on deep learning, in particular to the field of artificial intelligence. Basic information of a driver and a passenger and mutual evaluation data between the driver and the passenger are collected and preprocessed, statistics is replaced and converted into standardized scoring digits, and scaling processing is carried out; dimensional influence and deviation among scoring digits can be eliminated, scoring results can be mapped to a more appropriate range, scores are more comparable, emotional tendency features are extracted, emotional polarity and keyword frequency features are expressed, key content and expression modes of texts can be known, and the method is suitable for popularization and application. A word embedding model is utilized to convert each vocabulary into a corresponding word vector, weighted averaging is carried out to obtain overall vector representation, an LSTM unit is selected to build a recurrent neural network model, and a total score of the driver and the conductor is output through a full connection layer. The emotional tendency and evaluation of drivers and passengers can be comprehensively considered.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and more specifically, to a method and device for a deep learning-based online car-hailing driver-passenger mutual evaluation system. Background Art

[0002] With the rapid development of mobile Internet, online ride-hailing has become an indispensable part of people's lives. Since drivers and passengers have different values ​​and living habits, how to establish a fair and just mutual evaluation system in this diversified context has become an important part of improving the quality of online ride-hailing services.

[0003] In traditional driver-passenger mutual evaluation, drivers can obtain passengers' ratings, but passengers cannot obtain real feedback from drivers, resulting in information asymmetry. Traditional star ratings often fail to truly reflect user satisfaction and are easily affected by a single unpleasant experience. Existing driver-passenger mutual evaluations also lack effective solutions to conflicts between drivers and passengers, resulting in problems that cannot be properly handled, affecting the driver and passenger experience. Summary of the invention

[0004] In view of the technical problems existing in the prior art, the present invention provides a method and device for an online car-hailing driver-passenger mutual evaluation system based on deep learning to solve the problems raised in the above background technology.

[0005] The technical solution of the present invention to solve the above technical problems is as follows: A method for a driver-passenger mutual evaluation system for online car-hailing based on deep learning, comprising the following steps:

[0006] S101. Collect and pre-process the basic information of the driver and the passengers themselves and the mutual evaluation data between the driver and the passengers, establish an initial vocabulary and merge and update it, use the updated unified vocabulary to replace the statistics and convert it into a standardized score number, and scale the standard score number;

[0007] S102. Obtain the basic information of the driver and the number of positive and negative words contained in the mutual evaluation data between the driver and the passenger, extract the sentiment tendency characteristics based on the comparison of the number of words and express the sentiment polarity, count the number of keywords and characters in the basic information of the driver and the passenger and the mutual evaluation data between the driver and the passenger, extract the keyword frequency characteristics, map the words to a continuous vector space using a word embedding model, convert each word into a corresponding word vector, and perform weighted average to obtain the overall vector representation;

[0008] S103. Output the sentiment classification label and corresponding sentiment analysis score of the feature, perform weighted average of the driver-passenger and passenger-driver dimensions according to the sentiment classification label and corresponding sentiment analysis score to obtain a comprehensive score, select the LSTM unit to build a recurrent neural network model and output the total score of the driver-passenger through the fully connected layer;

[0009] S104. Integrate the recurrent neural network model into the driver-passenger mutual evaluation system through server deployment and set up a data transmission interface, and obtain corresponding suggestions and matching measures through the real-time total scores of both drivers and passengers;

[0010] In a preferred embodiment, in S101, basic information of the driver and passengers themselves and mutual evaluation data between the driver and passengers are collected, missing values ​​of the basic information of the driver and passengers themselves and the mutual evaluation data between the driver and passengers are detected, and the records of the corresponding basic information and mutual evaluation data are deleted for the missing values, duplicate records of the basic information of the driver and passengers themselves and the mutual evaluation data between the driver and passengers are detected, and duplicate records are deleted, text processing is performed on the basic information of the driver and passengers themselves and the mutual evaluation data between the driver and passengers, an initial vocabulary is established, and synonyms and near synonyms are found through WordNet, for English text, words of different word forms are merged into the same root using stem extraction, for Chinese text, the text is segmented using a word segmentation tool, and the synonyms and near synonyms in the word segmentation results are merged, the vocabulary is updated, and the updated unified vocabulary is used for replacement, the original score number range is determined, and the statistics are converted into standardized score numbers, and the specific formula is:

[0011]

[0012] Among them, Score S Indicates a standardized score number, Score O Represents the original score number, Mean represents the average of the original score numbers, Deviation S The standard deviation of the original rating numbers is expressed to determine the new rating number range and scale the standard rating numbers. The specific formula is:

[0013] Score d =Score S *Range n +Min n

[0014] Among them, Score d Indicates converting the standardized rating number into a rating number within a new rating range. n Indicates the length of the new scoring range, Min n Indicates the minimum value of the range of scaled rating numbers.

[0015] In a preferred embodiment, in S102, the basic information of the driver and the passengers themselves and the mutual evaluation data between the drivers and the passengers are traversed using the sentiment dictionary, and the number of positive and negative words contained in the basic information of the driver and the passengers themselves and the mutual evaluation data between the drivers and the passengers are obtained. The sentiment tendency characteristics are extracted and the sentiment polarity is expressed based on the comparison of the number of words. The number of keywords and characters of the basic information of the driver and the passengers themselves and the mutual evaluation data between the drivers and the passengers are counted and the keyword frequency characteristics are extracted. The word embedding model is selected and the basic information of the driver and the passengers themselves and the mutual evaluation data between the drivers and the passengers are input for training, wherein the word embedding model maps words to a continuous vector space, converts each word into a corresponding word vector and performs weighted averaging to obtain the overall vector representation.

[0016] In a preferred embodiment, in S103, the two dimensions of driver and passenger are divided and the StructBERT model is selected, the sentiment tendency feature and the keyword frequency feature are input and the sentiment polarity of the feature is returned through the sentiment dictionary, keywords and the number of characters, the StructBERT model is loaded using the Transformers library and set to evaluation mode, the input is converted into a coding sequence using the tokenizer statement and passed to the StructBERT model, the sentiment classification label of the feature and the corresponding sentiment analysis score are output, and the two dimensions of driver and passenger and passenger are weighted averaged according to the sentiment classification label and the corresponding sentiment analysis score to obtain a comprehensive score, and the specific formula is:

[0017]

[0018] Among them, C z Indicates the comprehensive score, C sc represents the driver-passenger sentiment analysis score, W s represents the driver weight, C cs represents the sentiment analysis score of the passenger car, W c represents the passenger weight.

[0019] Furthermore, LSTM units are selected to build a recurrent neural network model. The accumulation speed of features is controlled by introducing a gating mechanism, wherein the gating mechanism includes an input gate, a forget gate, and an output gate. The input gate is divided into a sentiment score input layer and a basic information input layer. According to the basic information of the driver and passenger themselves input at the current time step, the mutual evaluation data between the driver and passenger, the comprehensive score, and the hidden state of the previous time step, the activation value of the input gate is calculated by the Sigmoid activation function and converted into a value between 0 and 1. The specific formula is:

[0020] R l =α(W l *[I t-1 ,x t ]+b l )

[0021] Among them, R l represents the activation value of the input gate, α() represents the Sigmoid activation function, W l represents the weight matrix, I t-1 represents the hidden state of the previous time step, x t represents the basic information of the driver and passenger, the mutual evaluation data between the driver and passenger, and the comprehensive score input at the current time step, b l Represents the bias term. The output gate selects the fully connected layer and calculates the activation value of the output gate through the Sigmoid activation function and converts it into a value between 0 and 1 according to the basic information of the driver and passenger input at the current time step, the mutual evaluation data between the driver and passenger and the comprehensive score, the hidden state of the previous time step, and the cell state of the current time step. The specific formula is:

[0022] R o =α(W o *[I t-1 ,x t ]+b o )

[0023] Among them, R o represents the activation value of the forget gate, α() represents the Sigmoid activation function, W o represents the weight matrix, I t-1 represents the hidden state of the previous time step, x t represents the basic information of the driver and passenger, the mutual evaluation data between the driver and passenger, and the comprehensive score input at the current time step, b o Represents the bias term, which outputs the total score of the driver and passenger through the fully connected layer.

[0024] In a preferred embodiment, in S104, the recurrent neural network model is integrated into the driver-passenger mutual evaluation system through server deployment and a data transmission interface is set, / api / feedback is used as the URL endpoint of the interface, and JSON is defined as the transmission format. The corresponding suggestions and matching measures are obtained through the real-time total scores of both the driver and the passenger, and a user interface is provided to allow users to view the evaluation results, suggestions and matching measures. A feedback function is provided, and a monitoring mechanism is set to monitor the evaluation data and the execution of suggestions in real time.

[0025] The present application also proposes a device for an online car-hailing driver-passenger mutual evaluation system based on deep learning, including: a driver-passenger mutual evaluation platform, a vocabulary, a StructBERT model and a Beidou navigation API, wherein the driver-passenger mutual evaluation platform is used to collect basic information of the driver and passengers themselves and mutual evaluation data between drivers and passengers, the vocabulary is used to find synonyms and near-synonyms through WordNet, the StructBERT model is used to obtain the comprehensive score given by the driver and passenger to each other in a certain order, and the Beidou navigation API is used to provide better route selection.

[0026] The beneficial effects of the present invention are as follows: statistical conversion into standardized scoring numbers and scaling processing help to eliminate dimensional effects and deviations between scoring numbers and map the scoring results to a more appropriate range, making the scoring more comparable; sentiment analysis feature extraction quantifies the emotional color in the text, helping to understand the emotional attitudes and tendencies between drivers and passengers; keyword frequency feature extraction helps to understand the key content and expression of the text, and better understand the communication and evaluation between drivers and passengers; word embedding models help to convert text data into numerical representations that can be processed by machine learning models; the emotional tendencies and evaluations of both drivers and passengers are comprehensively considered through the StructBERT model combination and the gated recurrent neural network model; server deployment and user interface instantly generate suggestions and matching measures, helping drivers and passengers improve service quality and experience, and facilitating drivers and passengers to provide feedback on evaluation results and suggestions. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a flow chart of the method of the present invention; DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0029] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, "plurality" means two or more, unless otherwise clearly and specifically defined.

[0030] In the description of the present application, the term "for example" is used to mean "used as an example, illustration or description". Any embodiment described as "for example" in the present application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid unnecessary details to obscure the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present application.

[0031] This embodiment provides Figure 1 A method and device for a deep learning-based online car-hailing driver-passenger mutual evaluation system is shown, which specifically includes the following steps:

[0032] S101. Collect and pre-process the basic information of the driver and the passengers themselves and the mutual evaluation data between the driver and the passengers, establish an initial vocabulary and merge and update it, use the updated unified vocabulary to replace the statistics and convert it into a standardized score number, and scale the standard score number;

[0033] Furthermore, the basic information of the drivers and passengers themselves and the mutual evaluation data between the drivers and passengers are collected, including name, gender, age, contact information, evaluation content, score and timestamp, the missing values ​​of the basic information of the drivers and passengers themselves and the mutual evaluation data between the drivers and passengers are detected, and the records of the corresponding basic information and mutual evaluation data are deleted for the missing values, the duplicate records of the basic information of the drivers and passengers themselves and the mutual evaluation data between the drivers and passengers are detected, and the duplicate records are deleted, the basic information of the drivers and passengers themselves and the mutual evaluation data between the drivers and passengers are processed for text to remove unnecessary punctuation and special characters, and only retain the text and the score digital content, establish an initial vocabulary and search for synonyms and near synonyms through WordNet, for English text, use stem extraction to merge words of different word forms into the same root, for Chinese text, use the word segmentation tool to segment the text, and merge the synonyms and near synonyms in the word segmentation results, update the vocabulary to ensure that the vocabulary contains all the merged words, use the updated unified vocabulary to replace, determine the original score number range, and statistically convert it into a standardized score number, the specific formula is:

[0034]

[0035] Among them, Score S Indicates a standardized score number, Score ORepresents the original score number, Mean represents the average of the original score numbers, Deviation S The standard deviation of the original rating numbers is expressed to determine the new rating number range and scale the standard rating numbers. The specific formula is:

[0036] Score d =Score S *Range n +Min n

[0037] Among them, Score d Indicates converting the standardized rating number into a rating number within a new rating range. n Indicates the length of the new scoring range, Min n Indicates the minimum value of the range of scaled rating numbers.

[0038] S102. Obtain the basic information of the driver and the number of positive and negative words contained in the mutual evaluation data between the driver and the passenger, extract the sentiment tendency characteristics based on the comparison of the number of words and express the sentiment polarity, count the number of keywords and characters in the basic information of the driver and the passenger and the mutual evaluation data between the driver and the passenger, extract the keyword frequency characteristics, map the words to a continuous vector space using a word embedding model, convert each word into a corresponding word vector, and perform weighted average to obtain the overall vector representation;

[0039] Furthermore, the sentiment dictionary is used to traverse the basic information of the driver and the passengers themselves and the mutual evaluation data between the driver and the passengers, and the number of positive and negative words contained in the basic information of the driver and the passengers themselves and the mutual evaluation data between the driver and the passengers are obtained. The sentiment tendency characteristics are extracted according to the comparison of the number of words and the sentiment polarity is expressed. When the number of positive words is greater than the number of negative words, the sentiment polarity is judged to be positive. When the number of positive words is less than the number of negative words, the sentiment polarity is judged to be negative. When the number of positive words is equal to the number of negative words, the sentiment polarity is judged to be neutral. The number of keywords and characters of the basic information of the driver and the passengers themselves and the mutual evaluation data between the driver and the passengers are counted and the keyword frequency characteristics are extracted. The word embedding model is selected and the basic information of the driver and the passengers themselves and the mutual evaluation data between the driver and the passengers are input for training. The word embedding model maps words to a continuous vector space, and is used for words with similar distances in the vector space. Each word is converted into a corresponding word vector and weighted average is performed to obtain the overall vector representation.

[0040] S103. Output the sentiment classification label and corresponding sentiment analysis score of the feature, perform weighted average of the driver-passenger and passenger-driver dimensions according to the sentiment classification label and corresponding sentiment analysis score to obtain a comprehensive score, select the LSTM unit to build a recurrent neural network model and output the total score of the driver-passenger through the fully connected layer;

[0041] Furthermore, the driver and passenger and passenger-driver dimensions are divided and the StructBERT model is selected to obtain the comprehensive score given by the driver and passenger to each other in a certain order. The sentiment tendency features and keyword frequency features are input and the sentiment polarity of the features is returned through the sentiment dictionary, keywords and number of characters. The StructBERT model is loaded using the Transformers library and set to evaluation mode. The input is converted into a coding sequence using the tokenizer statement and passed to the StructBERT model. The sentiment classification label and corresponding sentiment analysis score of the feature are output. The comprehensive score is obtained by weighted average of the driver and passenger and passenger-driver dimensions according to the sentiment classification labels and corresponding sentiment analysis scores. The specific formula is as follows:

[0042]

[0043] Among them, C z Indicates the comprehensive score, C sc represents the driver-passenger sentiment analysis score, W s represents the driver weight, C cs represents the sentiment analysis score of the passenger car, W c represents the passenger weight.

[0044] Furthermore, LSTM units are selected to build a recurrent neural network model. The accumulation speed of features is controlled by introducing a gating mechanism, including selectively adding new features and selectively forgetting previously accumulated features. The gating mechanism includes an input gate, a forget gate, and an output gate. The input gate is divided into a sentiment score input layer and a basic information input layer. According to the basic information of the driver and passenger themselves input at the current time step, the mutual evaluation data between the driver and passenger, the comprehensive score, and the hidden state of the previous time step, an LSTM layer is added to the output of the first layer of LSTM input gate to capture the time series at a deeper level. The activation value of the input gate is calculated by the Sigmoid activation function and converted into a value between 0 and 1. The specific formula is:

[0045] R l =α(W l *[I t-1 ,x t ]+b l )

[0046] Among them, R l represents the activation value of the input gate, α() represents the Sigmoid activation function, W l represents the weight matrix, I t-1 represents the hidden state of the previous time step, x t represents the basic information of the driver and passenger, the mutual evaluation data between the driver and passenger, and the comprehensive score input at the current time step, b lRepresents the bias term. The output gate selects the fully connected layer and calculates the activation value of the output gate through the Sigmoid activation function and converts it into a value between 0 and 1 according to the basic information of the driver and passenger input at the current time step, the mutual evaluation data between the driver and passenger and the comprehensive score, the hidden state of the previous time step, and the cell state of the current time step. The specific formula is:

[0047] R o =α(W o *[I t-1 ,x t ]+b o )

[0048] Among them, R o represents the activation value of the forget gate, α() represents the Sigmoid activation function, W o represents the weight matrix, I t-1 represents the hidden state of the previous time step, x t represents the basic information of the driver and passenger, the mutual evaluation data between the driver and passenger, and the comprehensive score input at the current time step, b o Represents the bias term, which outputs the total score of the driver and passenger through the fully connected layer.

[0049] S104. Integrate the recurrent neural network model into the driver-passenger mutual evaluation system through server deployment and set up a data transmission interface, and obtain corresponding suggestions and matching measures through the real-time total scores of both drivers and passengers;

[0050] Furthermore, the recurrent neural network model is integrated into the driver-passenger mutual evaluation system through server deployment and a data transmission interface is set up. / api / feedback is used as the URL endpoint of the interface, and JSON is defined as the transmission format. The corresponding suggestions and matching measures are obtained through the real-time total score of both the driver and the passengers. When multiple passengers complain that the driver’s service attitude is not good, a training plan to improve the driver’s service attitude is proposed. When passengers complain that the navigation route is inaccurate, the Beidou navigation API is connected to provide better route options. A user interface is provided to allow users to view evaluation results, suggestions and matching measures, and a feedback function is provided. A monitoring mechanism is set up to monitor the evaluation data and the execution of suggestions in real time.

[0051] An embodiment of the present invention also provides a device for an online car-hailing driver-passenger mutual evaluation system based on deep learning, including: a driver-passenger mutual evaluation platform, a vocabulary, a StructBERT model and a Beidou navigation API, wherein the driver-passenger mutual evaluation platform is used to collect basic information of the driver and passengers themselves and mutual evaluation data between drivers and passengers, the vocabulary is used to find synonyms and near-synonyms through WordNet, the StructBERT model is used to obtain the comprehensive scores given by the driver and passengers to each other in a certain order, and the Beidou navigation API is used to provide better route selection.

[0052] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0053] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented 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.

[0054] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0055] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0056] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0057] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0058] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for a deep learning-based online car-hailing driver-passenger mutual evaluation system, characterized in that: The specific steps include: S101. Collect and pre-process the basic information of the driver and the passengers themselves and the mutual evaluation data between the driver and the passengers, establish an initial vocabulary and merge and update it, use the updated unified vocabulary to replace the statistics and convert it into a standardized score number, and scale the standard score number; S102. Obtain the basic information of the driver and the number of positive and negative words contained in the mutual evaluation data between the driver and the passenger, extract the sentiment tendency characteristics based on the comparison of the number of words and express the sentiment polarity, count the number of keywords and characters in the basic information of the driver and the passenger and the mutual evaluation data between the driver and the passenger, extract the keyword frequency characteristics, map the words to a continuous vector space using a word embedding model, convert each word into a corresponding word vector, and perform weighted average to obtain the overall vector representation; S103. Output the sentiment classification label and corresponding sentiment analysis score of the feature, perform weighted average of the driver-passenger and passenger-driver dimensions according to the sentiment classification label and corresponding sentiment analysis score to obtain a comprehensive score, select the LSTM unit to build a recurrent neural network model and output the total score of the driver-passenger through the fully connected layer; S104. Integrate the recurrent neural network model into the driver-passenger mutual evaluation system through server deployment and set up a data transmission interface to obtain corresponding suggestions and matching measures through the real-time total scores of both the driver and the passenger.

2. According to the method of the deep learning-based online car-hailing driver-passenger mutual evaluation system of claim 1, it is characterized by: In S101, the specific steps of preprocessing are: detecting missing values ​​of the basic information of the driver and passengers themselves and the mutual evaluation data between the driver and passengers, deleting the records of the corresponding basic information and mutual evaluation data for the missing values, detecting duplicate records of the basic information of the driver and passengers themselves and the mutual evaluation data between the driver and passengers, and deleting the duplicate records, and performing text processing on the basic information of the driver and passengers themselves and the mutual evaluation data between the driver and passengers.

3. The method of the deep learning-based online car-hailing driver-passenger mutual evaluation system according to claim 2 is characterized by: The specific steps of text processing are: establishing an initial vocabulary and searching for synonyms and near-synonyms through WordNet, updating the vocabulary, replacing with the updated unified vocabulary, determining the original score number range, and statistically converting it into a standardized score number, determining a new score number range, and scaling the standardized score number.

4. The method of the deep learning-based online car-hailing driver-passenger mutual evaluation system according to claim 3 is characterized by: The specific formula for converting the statistics into standardized scoring numbers is: Among them, Score S Indicates a standardized score number, Score O Represents the original score number, Mean represents the average of the original score numbers, Deviation S Represents the standard deviation of the original score number. The specific formula for the scaling process is: Score d =Score S *Range n +Min n Among them, Score d Indicates converting the standardized rating number into a rating number within a new rating range. n Indicates the length of the new scoring range, Min n Indicates the minimum value of the range of scaled rating numbers.

5. The method of the deep learning-based online car-hailing driver-passenger mutual evaluation system according to claim 1 is characterized by: In S102, the basic information of the driver and the passenger themselves and the number of positive and negative words contained in the mutual evaluation data between the driver and the passengers are obtained, the sentiment tendency characteristics are extracted and the sentiment polarity is expressed based on the comparison of the number of words, the number of keywords and characters of the basic information of the driver and the passenger themselves and the mutual evaluation data between the driver and the passengers are counted and the keyword frequency characteristics are extracted, the word embedding model is selected and the basic information of the driver and the passenger themselves and the mutual evaluation data between the driver and the passengers are input for training, each word is converted into a corresponding word vector and weighted average is performed to obtain the overall vector representation.

6. The method of the deep learning-based online car-hailing driver-passenger mutual evaluation system according to claim 1, characterized in that: In S103, the two dimensions of driver and passenger and passenger-driver are divided and the StructBERT model is selected, the sentiment tendency feature and the keyword frequency feature are input and the sentiment polarity of the feature is returned through the sentiment dictionary, keywords and the number of characters, the StructBERT model is loaded using the Transformers library and set to evaluation mode, the input is converted into a coding sequence using a tokenizer statement and passed to the StructBERT model, the sentiment classification label of the feature and the corresponding sentiment analysis score are output, and the two dimensions of driver and passenger and passenger-driver are weighted averaged according to the sentiment classification label and the corresponding sentiment analysis score to obtain a comprehensive score.

7. The method of the deep learning-based online car-hailing driver-passenger mutual evaluation system according to claim 6, characterized in that: The specific formula for obtaining the comprehensive score by weighted average is: Among them, C z Indicates the comprehensive score, C sc represents the driver-passenger sentiment analysis score, W s represents the driver weight, C cs represents the sentiment analysis score of the passenger car, W c represents the passenger weight.

8. The method of the deep learning-based online car-hailing driver-passenger mutual evaluation system according to claim 1, characterized in that: In S103, the specific steps of selecting LSTM units to build a recurrent neural network model and outputting the total score of both the driver and the passenger through the fully connected layer are: by introducing a gating mechanism to control the accumulation speed of features, the input gate is divided into an emotional score input layer and a basic information input layer, and according to the basic information of the driver and the passenger themselves input at the current time step, the mutual evaluation data between the driver and the passenger, the comprehensive score and the hidden state of the previous time step, the activation value of the input gate is calculated by the Sigmoid activation function and converted into a value between 0 and 1. The specific formula is: R l =α(W l *[I t-1 ,x t ]+b l ) Among them, R l represents the activation value of the input gate, α() represents the Sigmoid activation function, W l represents the weight matrix, I t-1 represents the hidden state of the previous time step, x t represents the basic information of the driver and passenger, the mutual evaluation data between the driver and passenger, and the comprehensive score input at the current time step, b l Represents the bias term. The output gate selects the fully connected layer and calculates the activation value of the output gate through the Sigmoid activation function and converts it into a value between 0 and 1 according to the basic information of the driver and passenger input at the current time step, the mutual evaluation data between the driver and passenger and the comprehensive score, the hidden state of the previous time step, and the cell state of the current time step. The specific formula is: R o =α(W o *[I t-1 ,x t ]+b o ) Among them, R o represents the activation value of the forget gate, α() represents the Sigmoid activation function, W o represents the weight matrix, I t-1 represents the hidden state of the previous time step, x t represents the basic information of the driver and passenger, the mutual evaluation data between the driver and passenger, and the comprehensive score input at the current time step, b o Represents the bias term.

9. The method of the deep learning-based online car-hailing driver-passenger mutual evaluation system according to claim 1, characterized in that: In S104, the recurrent neural network model is integrated into the driver-passenger mutual evaluation system through server deployment and a data transmission interface is set. / api / feedback is used as the URL endpoint of the interface. The corresponding suggestions and matching measures are obtained through the real-time total scores of both the driver and the passenger. A user interface and feedback function are provided and a monitoring mechanism is set to monitor the evaluation data and the execution of the suggestions in real time.

10. A device for a deep learning-based online car-hailing driver-passenger mutual evaluation system, characterized in that: The device includes: The driver-passenger mutual evaluation platform, vocabulary, StructBERT model and instructions in Beidou navigation API are used to implement a method of a deep learning-based online car-hailing driver-passenger mutual evaluation system according to any one of claims 1-9.

Images