An AI-based intelligent hotel room service evaluation method

Abstract

The present invention discloses an AI-based intelligent hotel room service evaluation and assessment method, which relates to the technical field of hotel room service evaluation and assessment, and includes the following steps: S1: Data collection: Collecting customer feedback data on room service by designing a questionnaire, deploying a feedback system, and integrating an application program interface (API). The AI-based intelligent hotel room service evaluation and assessment method enables the hotel to have an in-depth understanding of the personalized needs and preferences of each customer through a detailed scoring system and multi-dimensional data collection; it can enhance customers' sense of identity and loyalty to the hotel, and make customers feel the hotel's attention and attention to them, thereby improving customers' overall experience and satisfaction, and promoting customers' repeated bookings and word-of-mouth communication.

Description

Technical Field

[0001] The present invention relates to the technical field of hotel room service evaluation and assessment, and in particular to an AI-based intelligent hotel room service evaluation and assessment method. Background Art

[0002] Room service is an important part of the hotel industry, which directly affects the customer's stay experience and satisfaction. Effective service evaluation can not only help hotels discover and solve problems in a timely manner, but also improve the overall service quality and enhance customer loyalty. Traditional room service evaluation mainly relies on customer feedback and managers' subjective judgment, but this method has many limitations. In recent years, with the development of AI technology, artificial intelligence technology has been increasingly used in various fields, especially in data analysis, natural language processing (NLP) and machine learning. Significant progress has been made. These technologies provide new solutions for room service evaluation.

[0003] However, existing evaluation methods still rely on a single data source, such as customer ratings or manager observations, and lack multi-dimensional data support; customer feedback may be subjective and emotional, and manager evaluations may also be biased, resulting in low data quality. Summary of the invention

[0004] In order to solve the above technical problems, the present invention is implemented by the following technical solutions: an AI-based intelligent hotel room service evaluation method, comprising the following steps:

[0005] S1: Data collection: Collect customer feedback on room service by designing questionnaires, deploying feedback systems, and integrating application programming interfaces (APIs);

[0006] S2: Data preprocessing: Data cleaning and feature extraction are performed on the collected feedback data to obtain feedback features, which include bed comfort score, temperature control score, noise level score, and intelligent experience score;

[0007] S3: Calculation of room experience score: Perform secondary cleaning based on feedback features and perform feature extraction to extract the room comfort score and intelligent experience score, and then perform weighted calculation on these two scores to obtain the room experience score;

[0008] S4: Calculation of customer satisfaction score: Based on the cleaning of feedback data, extract from customer satisfaction data to obtain customer service score, emotion score, length of stay, front desk service score, catering service score and entertainment facility score, and perform weighted calculation to obtain customer satisfaction score;

[0009] S5: Comprehensive evaluation: The room experience score and customer satisfaction score are weighted to obtain a comprehensive evaluation score, and the comprehensive evaluation score is normalized to a range of 0-100;

[0010] S6: Model training and optimization: Prepare training data sets, perform model training and verification through the selected machine learning model, deploy the trained model to the production environment, and achieve real-time evaluation;

[0011] S7: Continuous Improvement: Regularly collect new customer feedback data, retrain the model and optimize the weight coefficients, and formulate improvement measures based on the evaluation results to improve service quality.

[0012] Preferably, the data collection step includes:

[0013] S11: Design a questionnaire covering room comfort, intelligent experience and customer satisfaction;

[0014] S12: Deploy feedback systems at the hotel front desk, in the guest rooms and on the website to facilitate customers to fill in the questionnaire;

[0015] S13: extract hotel business data from the hotel management system using an application programming interface (API);

[0016] S14: The room comfort data, intelligent experience data and customer satisfaction data after the customer checks in are collected as feedback data.

[0017] Among them, hotel business data includes customer information, customer stay duration, reservation management, financial management, room status, and service requests.

[0018] Preferably, the data preprocessing step includes:

[0019] S21: Clean the collected feedback data to remove invalid or abnormal data;

[0020] S22: Perform feature extraction based on the cleaned feedback data, extract the bed comfort score, temperature control score, noise level score, and intelligent experience score in the feedback data, and obtain feedback features.

[0021] Preferably, in the room experience score calculation step in S3:

[0022] The room comfort score and the intelligent experience score are weighted to obtain the room experience score. The calculation formula of the room experience score E is:

[0023] ;

[0024] in, : Customer rating of room comfort;

[0025] : It is the customer's rating of the intelligent facilities in the room;

[0026] w1 is the weight coefficient of the customer's score on the room comfort, w2 is the weight coefficient of the customer's score on the room's intelligent facilities; and w1+w2=1.

[0027] Preferably, the room comfort score includes a bed comfort score, a temperature control score, and a noise level score, and the customer's room comfort score ranges from 0 to 100, with 100 representing the most comfortable;

[0028] The intelligent experience score is the customer's score on the room's intelligent facilities, including smart lights, smart curtains, smart temperature control and smart speakers. The customer's score range for the room's intelligent facilities is 0-100, with 100 indicating the best intelligent experience.

[0029] Preferably, the room comfort score is calculated as follows:

[0030] Room comfort score calculation formula: ;

[0031] in, : Bed comfort rating; : Temperature control score; : Noise level score; ; : Weight coefficient of bed comfort score; : Weight coefficient of temperature control score; : Weight coefficient for noise level score.

[0032] Preferably, the intelligent experience score is calculated as follows:

[0033] Intelligent experience score calculation formula:

[0034] ;in, : Smart lighting experience rating, : Smart curtain experience rating;

[0035] : Intelligent temperature control experience rating, : Smart speaker experience rating;

[0036] ; : Weight coefficient of smart lighting experience score; : Weight coefficient of smart curtain experience score; : Weight coefficient of smart temperature control experience score; : Weight coefficient of smart speaker experience score.

[0037] Preferably, in the customer satisfaction score calculation step, the customer satisfaction score is obtained by weighted calculation of the customer service score, the emotion score, the length of stay, the front desk service score, the catering service score and the entertainment facility score;

[0038] Among them, calculate the customer satisfaction score The calculation formula is:

[0039] ;

[0040] : Room service rating, where customers rate room service on a scale of 1-5 stars;

[0041] : Sentiment score;

[0042] : The length of time the customer stays in the hotel, in days;

[0043] : Customers’ ratings of front desk services, with the ratings ranging from 1 to 5 stars;

[0044] : Customer ratings of catering services;

[0045] : Customer ratings of recreational facilities, ranging from 1 to 5 stars;

[0046] ;

[0047] in, is the weight coefficient of the corresponding score above.

[0048] Preferably, the sentiment score is calculated as follows:

[0049] Use the pre-trained sentiment analysis model f to perform sentiment analysis on the text review T and output the sentiment score ; The sentiment score calculation formula is as follows: ;

[0050] in, The value range is [-1,1], where -1 means very negative and 1 means very positive.

[0051] Preferably, in the comprehensive evaluation step, the room experience score and the customer satisfaction score are weighted and calculated, and the comprehensive evaluation score is normalized to a range of 0-100 by normalization processing, wherein the formula for the normalization processing is:

[0052] ;

[0053] in, is the normalized comprehensive evaluation score;

[0054] is the comprehensive assessment score; is the maximum comprehensive evaluation score;

[0055] Among them, the comprehensive evaluation score Calculated by the following formula:

[0056] ;

[0057] in, ; is the weight coefficient of the room experience score; is the weight coefficient of customer satisfaction score.

[0058] Preferably, in the model training and optimization step, a suitable machine learning model is selected based on linear regression, decision tree, and random forest, the selected model is trained and verified, and deployed to the production environment to achieve real-time evaluation;

[0059] The mean square error (MSE) is used as the loss function, and the goal is to minimize the actual score. and predicted scores The difference between them is calculated as follows: ;

[0060] Use the gradient descent method to adjust the weight coefficient so that the loss function gradually decreases. The calculation formula is as follows: ;

[0061] in, : is the learning rate; : is the loss function for weights The partial derivative of .

[0062] The present invention provides an AI-based intelligent hotel room service evaluation method, which has the following beneficial effects:

[0063] This AI-based smart hotel room service evaluation and assessment method enables hotels to gain in-depth understanding of the personalized needs and preferences of each customer through a detailed scoring system and multi-dimensional data collection; including providing personalized lighting scene settings for customers who have high requirements for smart lighting by analyzing customers' scores on different intelligent facilities; providing targeted care and compensation measures for customers with special needs or who have experienced poor experiences based on customer feedback in emotional scores and text reviews; this personalized service can enhance customers' identification and loyalty to the hotel, making customers feel the hotel's attention and attention to them, thereby improving customers' overall experience and satisfaction, and promoting customers' repeat bookings and word-of-mouth communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 The present invention is a flowchart of an AI-based intelligent hotel room service evaluation method. DETAILED DESCRIPTION

[0065] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0066] like Figure 1 As shown, the present invention provides a technical solution: an AI-based intelligent hotel room service evaluation method, comprising the following steps:

[0067] S1: Data collection: Collect customer feedback on room service by designing questionnaires, deploying feedback systems, and integrating application programming interfaces (APIs);

[0068] S2: Data preprocessing: Data cleaning and feature extraction are performed on the collected feedback data to obtain feedback features, which include bed comfort score, temperature control score, noise level score, and intelligent experience score;

[0069] S3: Calculation of room experience score: Perform secondary cleaning based on feedback features and perform feature extraction to extract the room comfort score and intelligent experience score, and then perform weighted calculation on these two scores to obtain the room experience score;

[0070] S4: Calculation of customer satisfaction score: Based on the cleaning of feedback data, extract from customer satisfaction data to obtain customer service score, emotion score, length of stay, front desk service score, catering service score and entertainment facility score, and perform weighted calculation to obtain customer satisfaction score;

[0071] S5: Comprehensive evaluation: The room experience score and customer satisfaction score are weighted to obtain a comprehensive evaluation score, and the comprehensive evaluation score is normalized to a range of 0-100;

[0072] S6: Model training and optimization: Prepare training data sets, perform model training and verification through the selected machine learning model, deploy the trained model to the production environment, and achieve real-time evaluation;

[0073] S7: Continuous Improvement: Regularly collect new customer feedback data, retrain the model and optimize the weight coefficients, and formulate improvement measures based on the evaluation results to improve service quality.

[0074] Comprehensive evaluation based on room experience and customer satisfaction; by systematically collecting and analyzing various relevant data, we can gain an in-depth and comprehensive understanding of customers' hotel experience, avoid the one-sidedness of single-factor evaluation, and provide hotels with a comprehensive and multi-angle perspective to examine their own service quality;

[0075] It not only focuses on the comfort of the guest room hardware facilities, but also takes into account the customer's interactive experience with each service link of the hotel during the entire stay, as well as the emotional tendencies expressed by the customer in text reviews, so as to accurately locate the advantages and disadvantages of the service, providing a solid data foundation for targeted improvement of services.

[0076] The data collection steps include:

[0077] S11: Design a questionnaire covering room comfort, intelligent experience and customer satisfaction;

[0078] S12: Deploy feedback systems at the hotel front desk, in the guest rooms and on the website to facilitate customers to fill in the questionnaire;

[0079] S13: extract hotel business data from the hotel management system using an application programming interface (API);

[0080] S14: The room comfort data, intelligent experience data and customer satisfaction data after the customer checks in are collected as feedback data.

[0081] Among them, hotel business data includes customer information, customer stay duration, reservation management, financial management, room status, and service requests.

[0082] The data preprocessing steps include:

[0083] S21: Clean the collected feedback data to remove invalid or abnormal data;

[0084] S22: Perform feature extraction based on the cleaned feedback data, extract the bed comfort score, temperature control score, noise level score, and intelligent experience score in the feedback data, and obtain feedback features.

[0085] In the room experience score calculation step in S3:

[0086] The room comfort score and the intelligent experience score are weighted to obtain the room experience score. The calculation formula of the room experience score E is:

[0087] ;

[0088] in, : Customer rating of room comfort;

[0089] : It is the customer's rating of the intelligent facilities in the room;

[0090] w1 is the weight coefficient of the customer's score on the room comfort, w2 is the weight coefficient of the customer's score on the room's intelligent facilities; and w1+w2=1.

[0091] The room comfort score includes the bed comfort score, temperature control score, and noise level score. Customers rate the room comfort on a scale of 0-100, with 100 indicating the most comfortable.

[0092] The intelligent experience score is the customer's score on the room's intelligent facilities, including smart lights, smart curtains, smart temperature control and smart speakers. The customer's score range for the room's intelligent facilities is 0-100, with 100 indicating the best intelligent experience.

[0093] To calculate the room comfort score, the process is as follows:

[0094] Room comfort score calculation formula: ;

[0095] in, : Bed comfort rating; : Temperature control score; : Noise level score; ; : Weight coefficient of bed comfort score; : Weight coefficient of temperature control score; : Weight coefficient for noise level score.

[0096] The process of calculating the intelligent experience score is as follows:

[0097] Intelligent experience score calculation formula:

[0098] ;in, : Smart lighting experience rating, : Smart curtain experience rating;

[0099] : Intelligent temperature control experience rating, : Smart speaker experience rating;

[0100] ; : Weight coefficient of smart lighting experience score; : Weight coefficient of smart curtain experience score; : Weight coefficient of smart temperature control experience score; : Weight coefficient of smart speaker experience score.

[0101] In the customer satisfaction score calculation step, the customer satisfaction score is obtained by weighted calculation of the customer service score, emotion score, length of stay, front desk service score, catering service score and entertainment facility score;

[0102] Among them, calculate the customer satisfaction score The calculation formula is:

[0103] ;

[0104] : Room service rating, where customers rate room service on a scale of 1-5 stars;

[0105] : Sentiment score;

[0106] : The length of time the customer stays in the hotel, in days;

[0107] : Customers’ ratings of front desk services, with the ratings ranging from 1 to 5 stars;

[0108] : Customer ratings of catering services;

[0109] : Customer ratings of recreational facilities, ranging from 1 to 5 stars;

[0110] ;

[0111] in, is the weight coefficient of the corresponding score above.

[0112] To calculate the sentiment score, the process is as follows:

[0113] Use the pre-trained sentiment analysis model f to perform sentiment analysis on the text review T and output the sentiment score ; The sentiment score calculation formula is as follows: ;

[0114] in, The value range is [-1,1], where -1 means very negative and 1 means very positive.

[0115] It should be further explained that in the specific implementation process, the complete process of calculating the sentiment score is as follows:

[0116] Data preparation: Collect text review data left by customers on the hotel's online booking platform and the hotel's official website review area; the text review data includes evaluation content on room service, hotel facilities, and employee attitudes, and contains rich customer sentiment information; pre-process the collected text review data, including removing punctuation and stop words, and converting the text to lowercase letters to improve the efficiency and accuracy of subsequent processing; at the same time, perform stem extraction and lemmatization on the text to restore the words to their basic form, including restoring "running" to "run" and "better" to "good", to reduce the diversity of vocabulary and facilitate subsequent analysis;

[0117] Select a pre-trained sentiment analysis model: Select a model suitable for the text characteristics of the hotel industry from the existing pre-trained sentiment analysis models, that is, the pre-trained model GPT based on the deep learning framework;

[0118] Model adaptation and fine-tuning: Based on the differences between the selected pre-trained model and the text data of the hotel industry, the model is adapted and fine-tuned; the model is further trained using a specific corpus in the hotel field to enable the model to better understand hotel-related vocabulary, context, and emotional expressions; including collecting a large number of positive and negative review texts from the hotel industry, conducting targeted training on the model, and adjusting the model's parameters to adapt to the task of sentiment analysis of hotel customer reviews;

[0119] During the fine-tuning process, the preprocessed hotel text review data is divided into training set, validation set and test set; the training set is used to train the model, and the validation set is used to adjust the learning rate and number of iterations in the model's hyperparameters; the test set is used to evaluate the performance of the model on unseen data; through multiple iterations of training and validation, the optimal model parameter configuration is found to improve the accuracy of the model's sentiment analysis of hotel texts;

[0120] Sentiment analysis and score calculation: Input the preprocessed hotel text review data into the fine-tuned pre-trained sentiment analysis model; Use the model to perform semantic understanding and sentiment classification on the text, and divide the sentiment into positive, negative and neutral categories; Calculate the sentiment score based on the output of the model; In the range of [-1,1], if the model directly outputs the sentiment score, directly obtain the score as the sentiment score of the customer review; Based on the sentiment category labels: "positive", "negative" and "neutral", if the model outputs the sentiment category label, a score value is pre-set for each category, including: positive sentiment corresponds to a score of 1, negative sentiment corresponds to a score of -1, and neutral sentiment corresponds to a score of 0; Then assign the corresponding score to the customer review according to the sentiment category predicted by the model;

[0121] Result verification and optimization: Use the test set to evaluate the performance of the model for calculating sentiment scores. Use the accuracy, recall, and F1 value in the evaluation indicators to measure the accuracy of the model. The accuracy rate indicates the proportion of samples correctly predicted by the model to the total number of samples, the recall rate indicates the proportion of samples correctly predicted to the actual number of samples, and the F1 value is an indicator that comprehensively considers the accuracy rate and the recall rate. By evaluating these indicators, we can understand the performance of the model in sentiment analysis and score calculation.

[0122] If the model performance does not meet expectations, the model is further optimized, including increasing the amount of training data, adjusting the model structure, and optimizing hyperparameters. At the same time, the text preprocessing method is improved, including trying different stem extraction algorithms or stop word lists to improve the model's understanding of text sentiment and the accuracy of score calculations. The above steps are repeated until the model performance meets the requirements, thereby obtaining accurate and reliable sentiment score calculation results, which provide strong support for subsequent customer satisfaction score calculations and hotel service improvements.

[0123] In the comprehensive evaluation step, the room experience score and customer satisfaction score are weighted and calculated, and the comprehensive evaluation score is normalized to a range of 0-100 through normalization. The normalization formula is:

[0124] ;

[0125] in, is the normalized comprehensive evaluation score;

[0126] is the comprehensive assessment score; is the maximum comprehensive evaluation score;

[0127] Among them, the comprehensive evaluation score Calculated by the following formula:

[0128] ;

[0129] in, ; is the weight coefficient of the room experience score; is the weight coefficient of customer satisfaction score.

[0130] In the model training and optimization step, a suitable machine learning model is selected based on linear regression, decision tree, and random forest, the selected model is trained and verified, and deployed to the production environment for real-time evaluation;

[0131] The process in the model training and optimization steps is as follows:

[0132] First, select a machine learning model: choose a suitable model from linear regression, decision tree, and random forest for subsequent training;

[0133] Then, prepare the training data set: use the data collected and processed in the previous steps as the training data set, including room experience score related data and customer satisfaction score related data;

[0134] Among them, the guest room experience score-related data includes bed comfort score, temperature control score, noise level score, intelligent experience score, and calculated guest room experience score;

[0135] Customer satisfaction score related data includes room service score, emotion score, length of stay, front desk service score, catering service score and entertainment facility score, as well as the calculated customer satisfaction score;

[0136] Finally, model training and validation:

[0137] The mean square error (MSE) is used as the loss function, and the goal is to minimize the actual score. and predicted scores The difference between them is calculated as follows:

[0138] ;

[0139] The gradient descent method is used to adjust the weight coefficient so that the loss function gradually decreases. The calculation formula is as follows:

[0140] ;

[0141] in, : is the learning rate; : is the loss function for weights The partial derivative of ; Through continuous iterative training and verification, the performance of the model is evaluated to ensure the accuracy and generalization ability of the model on the training data;

[0142] Deploy to production environment: Deploy trained and verified models to the production environment for real-time evaluation so that new customer feedback data can be processed and evaluated.

[0143] It should be further explained that in the specific implementation process, the continuous improvement steps are as follows:

[0144] S71: Regularly collect new customer feedback data: Continue to apply the method used in the data collection step S1, that is, through the designed questionnaire containing room comfort, intelligent experience and customer satisfaction related content, the feedback system deployed at the hotel front desk, in the guest room and on the website is convenient for customers to fill in the questionnaire, and use the application program interface API to extract hotel business data from the hotel management system, including customer information, customer stay time, reservation management, financial management, room status, service request, to obtain new room comfort data, intelligent experience data and customer satisfaction data after the customer checks in, and collect them as new feedback data;

[0145] S72: Retrain the model and optimize the weight coefficients:

[0146] S721: Based on step S2, the newly collected feedback data is preprocessed, including data cleaning, removing invalid or abnormal data, and feature extraction, extracting feedback features of bed comfort score, temperature control score, noise level score, and intelligent experience score;

[0147] S722: Based on step S3, the room experience score is recalculated according to the processed data, that is, the room comfort score and the intelligent experience score are first weighted to obtain the room experience score, and the calculation process may involve adjustment and optimization of the weight coefficient to better reflect the customer's evaluation of the room experience;

[0148] Among them, the room comfort score is calculated by weighted calculation of the bed comfort score, temperature control score, and noise level score; the intelligent experience score is calculated by weighted calculation of the scores of smart lighting, smart curtains, smart temperature control, and smart speakers;

[0149] S723: Based on step S4, the customer satisfaction score is recalculated, and a new customer satisfaction score is obtained by weighted calculation of the room service score, sentiment score, length of stay, front desk service score, catering service score, and entertainment facility score in the new data. In this process, the weight coefficient corresponding to each score may also be optimized to make the customer satisfaction score more accurate; wherein, the sentiment score is obtained by analyzing the new text comments using a pre-trained sentiment analysis model;

[0150] S724: Based on step S5, a comprehensive evaluation is performed, and the newly calculated room experience score and customer satisfaction score are weighted again to obtain a comprehensive evaluation score, and the score is normalized to a range of 0-100 through normalization processing;

[0151] S725: Based on the new comprehensive evaluation results, the mean square error (MSE) is used as the loss function, and the weight coefficient is adjusted by the gradient descent method to retrain the model to improve the model's adaptability to new data and the evaluation accuracy;

[0152] S73: Develop improvement measures based on the evaluation results to improve service quality:

[0153] S731: In-depth analysis of the new evaluation results to identify problems and deficiencies in room service and customer satisfaction; if the bed comfort score in the room experience score is low, it means that the mattress and bedding need to be checked and replaced; if the smart experience score is not high, the smart device needs to be maintained, upgraded or the operation guide needs to be optimized;

[0154] S732: Formulate specific improvement measures based on the analyzed problems, including upgrading the hotel's facilities and equipment, optimizing service processes, and strengthening employee training, so as to improve the hotel's service quality and thus increase customer satisfaction;

[0155] S733: Continuously monitor the implementation effect of improvement measures, observe whether each scoring indicator has been improved, and evaluate whether the improvement measures are effective. If the expected effect is not achieved, further analyze the reasons, adjust the improvement measures or weight coefficients, and conduct model training and evaluation again to form a continuous cycle of improvement process to ensure the continuous improvement of hotel service quality.

[0156] It should be further explained that, in the specific implementation process, it has a multi-dimensional data collection function, can form a comprehensive evaluation system, and improve the accuracy and reliability of the data through data cleaning and preprocessing; by introducing specific quantitative indicators, including scores, sentiment analysis results, and task completion time, it ensures the objectivity and consistency of the evaluation, publicizes the evaluation standards and methods, and enhances the trust and acceptance of customers and employees.

[0157] It should be further explained that in the specific implementation process, with the help of model training and optimization and continuous improvement mechanism, the hotel can deploy the evaluation model in the actual operation environment to achieve real-time evaluation; this means that the hotel can obtain customer feedback immediately and promptly discover the emergence of service problems, such as multiple customers giving low ratings to the room noise level in a certain period of time, and the hotel can respond quickly;

[0158] Regularly collect new customer feedback data and retrain the model, and continuously optimize the weight coefficients so that the evaluation system can keep pace with the times and adapt to market changes and the dynamic evolution of customer needs; hotels can continuously adjust service strategies, improve facilities and equipment, optimize employee training, etc. based on the latest evaluation results to ensure that service quality always remains at a high level or even continuously improves, thereby enhancing the hotel's competitiveness in the market.

[0159] The entire evaluation and assessment method is based on the collection, processing and analysis of a large amount of objective data. Management can make decisions based on these accurate and detailed data, rather than relying solely on subjective experience or intuition. When deciding to upgrade guest room facilities, the management can refer to the weights and actual scores of each factor in the guest room experience score to determine priority improvement projects. In terms of personnel training, the training content and focus are arranged in a targeted manner based on the performance of customer service and front desk service in the customer satisfaction score. This data-driven decision-making method greatly improves the scientificity and rationality of decision-making, reduces decision-making risks, and enables hotel resources to be more effectively allocated to areas that are most in need of improvement and enhancement, thereby improving resource utilization efficiency and achieving refined management of hotel operations.

[0160] By analyzing the scores of each item, the hotel can identify which service links or facilities have a greater impact on customer satisfaction, and thus allocate resources reasonably. For example, if the intelligent experience score has a significant impact on the overall evaluation score, the hotel can increase investment in the update and maintenance of intelligent equipment and the research and development of related technologies. If the catering service score is low and the weight is high, more funds can be used for food procurement, chef training or restaurant environment improvement.

[0161] It avoids blind investment and waste of resources, and ensures that the hotel invests resources in key areas that can best improve customer satisfaction and hotel competitiveness within a limited budget, thereby maximizing input-output.

[0162] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. An AI-based intelligent hotel room service evaluation method, characterized in that: The following steps are involved: S1: Data collection: Collect customer feedback on room service by designing questionnaires, deploying feedback systems, and integrating application programming interfaces (APIs); S2: Data preprocessing: Data cleaning and feature extraction are performed on the collected feedback data to obtain feedback features, which include bed comfort score, temperature control score, noise level score, and intelligent experience score; S3: Calculation of room experience score: Perform secondary cleaning based on feedback features and perform feature extraction to extract the room comfort score and intelligent experience score, and then perform weighted calculation on these two scores to obtain the room experience score; S4: Calculation of customer satisfaction score: Based on the cleaning of feedback data, extract from customer satisfaction data to obtain room service score, emotion score, length of stay, front desk service score, catering service score and entertainment facility score, and perform weighted calculation to obtain customer satisfaction score; S5: Comprehensive evaluation: The room experience score and customer satisfaction score are weighted to obtain a comprehensive evaluation score, and the comprehensive evaluation score is normalized to a range of 0-100; S6: Model training and optimization: Prepare training data sets, perform model training and verification through the selected machine learning model, deploy the trained model to the production environment, and achieve real-time evaluation; S7: Continuous improvement: Regularly collect new customer feedback data, retrain the model and optimize the weight coefficients, and formulate improvement measures based on the evaluation results to improve service quality; The process of calculating the intelligent experience score is as follows: Intelligent experience score calculation formula: ;in, : Smart lighting experience rating, : Smart curtain experience rating; : Intelligent temperature control experience rating, : Smart speaker experience rating; ; : Weight coefficient of smart lighting experience score; : Weight coefficient of smart curtain experience score; : Weight coefficient of smart temperature control experience score; : Weight coefficient of smart speaker experience score.

2. According to claim 1, an AI-based intelligent hotel room service evaluation and assessment method is characterized in that: The data collection step includes: S11: Design a questionnaire covering room comfort, intelligent experience and customer satisfaction; S12: Deploy feedback systems at the hotel front desk, in the guest rooms and on the website to facilitate customers to fill in the questionnaire; S13: extract hotel business data from the hotel management system using an application programming interface (API); S14: The room comfort data, intelligent experience data and customer satisfaction data after the customer checks in are collected as feedback data.

3. According to claim 2, an AI-based intelligent hotel room service evaluation and assessment method is characterized in that: The data preprocessing step includes: S21: Clean the collected feedback data to remove invalid or abnormal data; S22: Perform feature extraction based on the cleaned feedback data, extract the bed comfort score, temperature control score, noise level score, and intelligent experience score in the feedback data, and obtain feedback features.

4. According to claim 3, an AI-based intelligent hotel room service evaluation and assessment method is characterized in that: In the room experience score calculation step in S3: The room comfort score and the intelligent experience score are weighted to obtain the room experience score. The calculation formula of the room experience score E is: ； in, : Customer rating of room comfort; : It is the customer's rating of the intelligent facilities in the room; w1 is the weight coefficient of the customer's score on the room comfort, w2 is the weight coefficient of the customer's score on the room's intelligent facilities; and w1+w2=1.

5. According to claim 4, an AI-based intelligent hotel room service evaluation and assessment method is characterized in that: The room comfort score includes the bed comfort score, temperature control score, and noise level score. The customer's room comfort score ranges from 0 to 100, with 100 indicating the most comfortable; The intelligent experience score is the customer's score on the room's intelligent facilities, including smart lights, smart curtains, smart temperature control and smart speakers. The customer's score range for the room's intelligent facilities is 0-100, with 100 indicating the best intelligent experience.

6. The AI-based intelligent hotel room service evaluation method according to claim 5 is characterized in that: To calculate the room comfort score, the process is as follows: Room comfort score calculation formula: ； in, : Bed comfort rating; : Temperature control score; : Noise level score; ; : Weight coefficient of bed comfort score; : Weight coefficient of temperature control score; : Weight coefficient for noise level score.

7. The AI-based intelligent hotel room service evaluation method according to claim 6 is characterized in that: In the customer satisfaction score calculation step, the customer satisfaction score is obtained by weighted calculation of the room service score, the emotion score, the length of stay, the front desk service score, the catering service score and the entertainment facility score; Among them, calculate the customer satisfaction score The calculation formula is: ； : Room service rating, where customers rate room service on a scale of 1-5 stars; : Sentiment score; : The length of time the customer stays in the hotel, in days; : Customers’ ratings of front desk services, with the ratings ranging from 1 to 5 stars; : Customer ratings of catering services; : Customer ratings of recreational facilities, ranging from 1 to 5 stars; ； in, is the weight coefficient of the corresponding score above.

8. The AI-based intelligent hotel room service evaluation method according to claim 7 is characterized in that: To calculate the sentiment score, the process is as follows: Use the pre-trained sentiment analysis model f to perform sentiment analysis on the text review T and output the sentiment score ; The sentiment score calculation formula is as follows: ; in, The value range is [-1,1], where -1 means very negative and 1 means very positive.

9. The AI-based intelligent hotel room service evaluation method according to claim 8, characterized in that: In the comprehensive evaluation step, the room experience score and the customer satisfaction score are weighted and calculated, and the comprehensive evaluation score is normalized to a range of 0-100 by normalization processing, wherein the formula for normalization processing is: ； in, is the normalized comprehensive evaluation score; is the comprehensive assessment score; is the maximum comprehensive evaluation score; Among them, the comprehensive evaluation score Calculated by the following formula: ； in, ; is the weight coefficient of the room experience score; is the weight coefficient of customer satisfaction score.

Images