Method for realizing hotel distribution intelligent order following based on enterprise private domain knowledge

Abstract

The invention belongs to the technical field of computers, and provides a method for realizing hotel distribution intelligent order following based on enterprise private domain knowledge, and the method comprises the following steps: obtaining server information of a hotel; performing dynamic cache frequency effect analysis on the server through the server information to determine a cache stable server; obtaining a target server by calculating a concurrent response time difference of the cache stabilization server; and generating and sending a personalized order following notification to the target user through the target server so as to send an operation instruction of order decision. According to the intelligent order following method provided by the embodiment of the invention, the load management of the server can be effectively optimized, the user experience can be improved, the stability and the anti-pressure capability of the system can be enhanced, and the platform can still operate efficiently and stably under the complicated conditions of high concurrency, hostile attack and the like.

Description

Technical Field

[0001] The present invention belongs to the field of computer technology, and particularly relates to a method for realizing intelligent order following in hotel distribution based on enterprise private domain knowledge. Background Art

[0002] Currently, in a hotel distribution platform, the load and response time of the server directly affect the user's access experience and the order decision-making process. However, when encountering malicious attacks (such as DoS / DDoS attacks, brute-force cracking) or improper requests (such as malicious crawlers), the cache may frequently expire and the hit rate may decrease, resulting in a significant increase in the server response time and even possible system crashes, seriously affecting the stability and availability of the system. Moreover, malicious attacks and crawlers exert pressure on the server by sending a large number of invalid requests, not only consuming valuable resources but also causing a large number of cache expirations, exacerbating the fluctuation of server performance, and thus affecting the user's access experience and the normal operation of the platform. Summary of the Invention

[0003] The present invention aims to solve at least one of the technical problems in the related art to some extent. To this end, the first object of the present invention is to propose a method for realizing intelligent order following in hotel distribution based on enterprise private domain knowledge, which can effectively optimize the load management of the server, improve the user experience, enhance the stability and anti-pressure ability of the system, and ensure that the platform can still operate efficiently and stably under complex conditions such as high concurrency and malicious attacks.

[0004] To achieve the above object, an embodiment of the first aspect of the present invention proposes a method for realizing intelligent order following in hotel distribution based on enterprise private domain knowledge, and the method includes the following steps:

[0005] S100, obtaining server information of a hotel;

[0006] S200, performing dynamic cache frequency analysis on the server through the server information to determine a cache-stable server;

[0007] S300, obtaining a target server by calculating the concurrent response time difference of the cache-stable server;

[0008] S400, generating and sending a personalized order following notice to a target user through the target server to send an operation instruction for order decision-making.

[0009] According to the intelligent order following method of the embodiment of the present invention, it can effectively optimize the load management of the server, improve the user experience, enhance the stability and anti-pressure ability of the system, and ensure that the platform can still operate efficiently and stably under complex conditions such as high concurrency and malicious attacks.

[0010] Further, in step S100, the server information of the hotel is obtained, where the hotel server information includes the response time of the server, the number of requests of the server, the concurrency of the server, and the cache hit rate;

[0011] In the hotel distribution platform, the load and response time of the server directly affect the user's access experience and the order decision-making process. However, in the face of malicious attacks (such as DoS / DDoS attacks, brute force cracking) or improper requests (such as malicious crawlers), the cache may frequently expire and the hit rate may decrease, resulting in a significant increase in the server response time, and even leading to system crashes, seriously affecting the stability and availability of the system. Moreover, malicious attacks and crawlers exert pressure on the server by sending a large number of invalid requests, which not only consume valuable resources but also cause a large number of cache expirations, exacerbating the fluctuation of server performance and thus affecting the user's access experience and the normal operation of the platform. To solve the above problems, the present invention proposes step S200.

[0012] Further, in step S200, the dynamic cache frequency analysis of the server is performed through the server information to determine the cache-stable server, including:

[0013] Let j represent the identifier of each server, j = 1, 2,..., K, where K represents the number of servers; denote the maximum cache hit rate of the j-th server in the most recent 1 hour as r(j), the minimum cache hit rate of the j-th server in the most recent 1 hour as p(j), and the average value of the minimum cache hit rates of all servers in the most recent 1 hour as Ms;

[0014] Perform dynamic cache frequency analysis on the server, where the steps for performing dynamic cache frequency analysis on the server include:

[0015] S201, calculate the cache frequency loss of the server in the most recent 1 hour;

[0016] Calculate the cache frequency loss ZG(j) of the j-th server in the most recent 1 hour, where ZG(j) is the difference between r(j) and p(j), and sum all ZG(j) and divide by K to obtain the stable cache frequency loss HDC;

[0017] Among them, the steady cache frequency loss HDC is used to quantify the changes and unevenness of the cache frequency loss of the server in the most recent hour and exclude the servers whose cache is invalidated and cache hit rate drops sharply due to DoS / DDoS attacks, brute force cracking, or malicious crawlers. In a Denial of Service (DoS) attack, the attacker consumes the cache of the target server by sending a large number of invalid requests, resulting in a large number of cache invalidations. Moreover, the behavior of malicious web crawlers is very difficult to accurately identify by existing technologies. Especially when these web crawlers simulate normal user behavior, they can disguise themselves as browsers to access, adopt a more dispersed and intermittent request method, making them almost indistinguishable from real user requests. In addition, the request content of web crawlers usually fully complies with the cache rules, but frequently requests non-existent or deleted pages, and this abnormal behavior is usually difficult to detect.

[0018] S202, Define an integer variable k, set its initial value to 1, and create a blank sequence ZRE;

[0019] S203, Conduct dynamic cache frequency efficiency analysis. Among them, the dynamic cache frequency efficiency analysis is as follows: Compare the values of ZG(j) and HDC, and compare the values of [p(j)+HDC] and Ms. If ZG(j) is less than HDC and [p(j)-HDC] is greater than Ms, then add ZG(j) to the sequence ZRE;

[0020] S204, Determine whether the dynamic cache frequency efficiency analysis is completed. The specific determination method is as follows: If the current variable k is less than K, then increment k by 1 and return to step S203 to continue the dynamic cache frequency efficiency analysis; If the current variable k is equal to K, it means that the dynamic cache frequency efficiency analysis has been processed and completed, and then go to step S205;

[0021] S205, Denote the servers corresponding to all elements in the sequence ZRE as cache-stable servers.

[0022] Among them, the dynamic cache frequency efficiency analysis is to analyze the cache hit rate of each server in the past hour, calculate its cache frequency loss, and evaluate the stability of the cache through these frequency loss values. It can not only reflect the changes in the server cache, but also identify cache invalidation problems caused by external malicious attacks or improper requests (such as DoS / DDoS attacks, brute force cracking, and malicious crawlers), so as to effectively distinguish unstable servers. By calculating and comparing the cache frequency loss and cache hit rate, servers affected by attacks or network crawlers are excluded. Especially in the case of DoS / DDoS attacks, brute force cracking, malicious crawlers, etc., these attacks and crawlers cause cache invalidation to the server through a large number of invalid requests. Through the quantitative method of the stable cache frequency loss HDC, it can accurately identify which servers have unstable caches due to abnormal requests, and then screen out servers with stable caches.

[0023] Among them, the cache frequency loss is used to reflect the fluctuation of the cache hit rate, and the stable cache frequency loss HDC is used to quantify the fluctuation and non-uniformity of the caches of all servers in the hotel, which can effectively exclude incorrect selections caused by abnormal fluctuations.

[0024] Beneficial effects of this step: Through the dynamic cache frequency efficiency analysis of the cache hit rate, it is possible to more accurately select servers with stable caches, avoiding system crashes or response delays caused by cache invalidation. Moreover, stable cache servers can still maintain a relatively low response time in the face of high concurrency or malicious attacks, avoiding delays and failed requests caused by server cache invalidation; by identifying and excluding cache invalidation servers caused by malicious attacks or crawlers, the fault tolerance of the hotel distribution platform in the face of malicious attacks is effectively enhanced. The system can share the load by selecting stable servers to prevent attackers or crawlers from affecting the normal operation of the platform.

[0025] Furthermore, in step S300, obtaining the target server by calculating the concurrent response time difference of cache-stable servers includes the following steps:

[0026] S301, calculate the concurrent response time difference of cache-stable servers;

[0027] Set an empty sequence STS as the set of response times of the cache stability server. The sequence takes the response time of the cache stability server as an element of the set STS. Use STS(i) to represent the average response time of the cache stability server with serial number i in the most recent 1 hour, where i represents the serial number of the cache stability server, and i = 1, 2, …, G, and G is the number of cache stability servers; obtain the shortest response time in STS and denote it as STSs; obtain the longest response time in STS and denote it as STSx; obtain the average value of all response times in STS as STSm. Set an empty sequence as the set KPT of the minimum concurrent numbers of the cache stability server in the most recent 1 hour. The sequence takes the minimum concurrent number of the cache stability server in the most recent 1 hour as an element of the set KPT. Use KPT(i) to represent the minimum concurrent number of the cache stability server with serial number i in the most recent 1 hour; set an empty sequence as the set KPU of the maximum concurrent numbers of the cache stability server in the most recent 1 hour. The sequence takes the maximum concurrent number of the cache stability server in the most recent 1 hour as an element of the set KPU. Use KPU(i) to represent the maximum concurrent number of the cache stability server with serial number i in the most recent 1 hour; obtain the average concurrent number in KPU and denote it as KPUs.

[0028] Among them, the minimum concurrent number of the cache stability server in the most recent 1 hour represents the lowest value of the concurrent request numbers received by the cache stability server in the past 1 hour. The maximum concurrent number of the cache stability server in the most recent 1 hour represents the highest value of the concurrent request numbers received by the cache stability server in the past 1 hour.

[0029] Among them, the concurrent response time difference includes the first concurrent response time difference and the second concurrent response time difference. Take the difference between STSx and STSm as the first concurrent response time difference SE; take the difference between STSm and STSs as the second concurrent response time difference SR.

[0030] S302. Calculate the concurrent load request duration of the cache stability server through the concurrent response time difference of the cache stability server.

[0031] Since the response time of the server is not only affected by the hardware performance, but also closely related to factors such as the type of requests, the workload of the system, and external dependencies. In a high-concurrency and high-load environment, the response time of the server usually shows non-linear fluctuations. Especially when the differences in the concurrent request volume and user behavior are large, it is impossible to predict whether the server can work properly only through the change of the response time itself; moreover, there are differences in the hardware configurations and load processing capabilities of different servers. Therefore, even under the same load conditions, the performance and response time of the servers may be different. Calculating the concurrent load request duration by combining the stability of the response time and the concurrent load capacity can more accurately understand the processing capabilities and load conditions of each server, so as to perform intelligent scheduling.

[0032] Further, an empty sequence POT is set to store the concurrent load request duration of the cache stability server. The sequence POT takes the concurrent load request duration POT(i) of the cache stability server as an element of the sequence POT, where i represents the serial number of the cache stability server, i = 1, 2, …, G, G is the number of cache stability servers, and POT(i) represents the concurrent load request duration of the cache stability server with the serial number i. The calculation method of the concurrent load request duration POT(i) is as follows:

[0033] POT(i) = STS(i) + |(SE - SR)| × (KPT(i) + KPU(i)) / KPUs;

[0034] Where, || is the absolute value symbol. The principle of the concurrent load request duration POT(i) is: based on the server's own response time STS(i), consider the fluctuation of the server response time through the concurrent response time difference (SE - SR): if SE is much larger than SR or much smaller than SR, it means that the maximum response time is far from the average value, while the minimum response time is close to the average value, indicating that the server's performance drops significantly under high load; if SE is close to SR, it means that the server's performance is relatively stable under high load and low load, and it is suitable as a target server. And consider the server's concurrent load capacity through (KPT(i) + KPU(i)) / KPUs: if the concurrent request numbers KPT(i) and KPU(i) of a certain server are both high, it means that the server is under greater pressure. If the server's concurrent load is low, the value of POT(i) is small, which means that the server can give priority to processing new requests.

[0035] S303. Determine the target server according to the concurrent load request duration of the cache stability server.

[0036] Specifically, sort the cache stability servers from small to large according to the size of the concurrent load request duration POT(i). The sequence number corresponding to the largest value in the load duration POT(i) is G, and the sequence number corresponding to the smallest value in the load duration POT(i) is 1. Preferentially select the cache stability server with a small sequence number as the target server according to the sequence number of the cache stability server.

[0037] The beneficial effects of this step are as follows: Through the calculation of POT(i), the system can dynamically evaluate the load conditions of each server, rather than relying solely on traditional metrics such as the current CPU usage rate and memory occupancy rate. By combining the stability of response time and concurrent load capacity, the POT(i) calculation method can effectively screen out high-performance and low-load servers, improving the rationality of request allocation. Since servers with larger POT(i) values usually indicate higher load or unstable response time, the system will preferentially select servers with smaller POT(i) values to process requests. By preferentially selecting servers with smaller POT(i) values, user requests can be processed faster, reducing page loading time, improving the smoothness of user access, and avoiding sudden delays or crashes caused by excessive server pressure. In the face of large-scale DoS / DDoS attacks, the calculation of POT(i) can effectively identify high-load servers and automatically allocate new requests to relatively stable servers, thereby reducing the impact of the attack on the entire system. And because web crawlers usually cause a decrease in the server's cache hit rate and an abnormal increase in the number of concurrent requests, the calculation of POT(i) can reflect these abnormal situations and avoid allocating critical business requests to servers that may be affected by crawlers.

[0038] S400. The target server generates and sends a personalized order tracking notice to the target user to send an operation instruction for order decision-making.

[0039] Specifically, in step S400, the target server analyzes the potential needs of the user based on the user's historical reservation behavior, browsing records, preference settings, and real-time market dynamics, and generates targeted order tracking notices, such as price change reminders, housing shortage warnings, and time-limited offer pushes. Further, the target server sends personalized order tracking notices to the target user through multiple channels (such as text messages, emails, or APP pushes), prompting the user to make an order decision as soon as possible, thereby improving the reservation conversion rate and optimizing the user experience.

[0040] By executing the method for realizing intelligent order tracking in hotel distribution based on enterprise private domain knowledge through the system for realizing intelligent order tracking in hotel distribution based on enterprise private domain knowledge, it can effectively optimize the load management of the server, improve the user experience, enhance the stability and anti-pressure ability of the system, and ensure that the platform can still operate efficiently and stably under complex situations such as high concurrency and malicious attacks. Brief Description of the Drawings

[0041] Figure 1 Shown is a flowchart of a method for realizing intelligent order tracking in hotel distribution based on enterprise private domain knowledge. Detailed Embodiments

[0042] Embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where like or similar reference numerals denote like or similar elements or elements having like or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present invention, and should not be construed as limiting the present invention.

[0043] Figure 1 Shown is a flowchart of a method for realizing intelligent follow-up of hotel distribution based on enterprise private domain knowledge.

[0044] Referring to Figure 1 , the present invention proposes a method for realizing intelligent follow-up of hotel distribution based on enterprise private domain knowledge, and the method includes the following steps:

[0045] S100, obtaining the server information of the hotel;

[0046] S200, performing dynamic cache frequency analysis on the server through the server information to determine the cache-stable server;

[0047] S300, obtaining the target server by calculating the concurrent response time difference of the cache-stable server;

[0048] S400, generating and sending a personalized follow-up notice to the target user through the target server to send an operation instruction for order decision-making.

[0049] According to the intelligent follow-up method of the embodiment of the present invention, it can effectively optimize the load management of the server, improve the user experience, enhance the stability and anti-pressure ability of the system, and ensure that the platform can still operate efficiently and stably under complex situations such as high concurrency and malicious attacks.

[0050] Further, in step S100, the server information of the hotel is obtained, where the hotel server information includes the response time of the server, the number of requests of the server, the concurrency of the server, and the cache hit rate;

[0051] Further, in step S200, performing dynamic cache frequency analysis on the server through the server information to determine the cache-stable server includes:

[0052] Let j represent the identifier of each server, j = 1, 2,..., K, where K represents the number of servers; denote the maximum cache hit rate of the jth server in the most recent 1 hour as r(j), the minimum cache hit rate of the jth server in the most recent 1 hour as p(j), and the average value of the minimum cache hit rates of all servers in the most recent 1 hour as Ms;

[0053] Performing dynamic cache frequency analysis on the server, where the steps of performing dynamic cache frequency analysis on the server include:

[0054] S201, Calculate the cache frequency loss of the server in the most recent 1 hour;

[0055] Calculate the cache frequency loss ZG(j) of the j-th server in the most recent 1 hour, where ZG(j) is the difference between r(j) and p(j), and add all ZG(j) and divide by K to obtain the stable cache frequency loss HDC;

[0056] S202, Define an integer variable k, set its initial value to 1, and create a blank sequence ZRE;

[0057] S203, Conduct dynamic cache frequency efficiency analysis. The dynamic cache frequency efficiency analysis is as follows: Compare the values of ZG(j) and HDC, and compare the values of [p(j) + HDC] and Ms. If ZG(j) is less than HDC and [p(j) - HDC] is greater than Ms, then add ZG(j) to the sequence ZRE;

[0058] S204, Determine whether the dynamic cache frequency efficiency analysis is completed. The specific determination method is as follows: If the current variable k is less than K, then increment k by 1 and return to step S203 to continue the dynamic cache frequency efficiency analysis; If the current variable k is equal to K, it means that the dynamic cache frequency efficiency analysis has been processed and completed, and then proceed to step S205;

[0059] S205, Denote the servers corresponding to all elements in the sequence ZRE as cache-stable servers.

[0060] Among them, the dynamic cache frequency efficiency analysis is to analyze the cache hit rate of each server in the past 1 hour, calculate its cache frequency loss, and evaluate the stability of the cache through these frequency loss values. It can not only reflect the changes in the server cache but also identify cache invalidation problems caused by external malicious attacks or improper requests (such as DoS / DDoS attacks, brute force attacks, and malicious crawlers), thereby effectively distinguishing unstable servers. By calculating and comparing the cache frequency loss and cache hit rate, servers affected by attacks or web crawlers are excluded. Especially in the cases of DoS / DDoS attacks, brute force attacks, malicious crawlers, etc., these attacks and crawlers cause cache invalidation to the server through a large number of invalid requests. Through the quantization method of the stable cache frequency loss HDC, it is possible to accurately identify which servers have unstable caches due to abnormal requests, and then screen out cache-stable servers.

[0061] Among them, the cache frequency loss is used to reflect the fluctuations in the cache hit rate, while the stable cache frequency loss HDC is used to quantify the fluctuations and non-uniformity of the caches of all the hotel's servers, and can effectively exclude incorrect selections caused by abnormal fluctuations.

[0062] Furthermore, in step S300, obtaining the target server by calculating the concurrent response time difference of the cache-stable servers includes the following steps:

[0063] S301, calculate the concurrent response time difference of the cache stability server;

[0064] Set an empty sequence STS as the response time set of the cache stability server. The sequence takes the response time of the cache stability server as an element of the set STS. Use STS(i) to represent the average response time of the cache stability server with serial number i in the most recent 1 hour, where i represents the serial number of the cache stability server, i = 1, 2, …, G, and G is the number of cache stability servers; obtain the shortest response time in STS and denote it as STSs; obtain the longest response time in STS and denote it as STSx; obtain the average value of all response times in STS as STSm. Set an empty sequence as the minimum concurrent number set KPT of the cache stability server in the most recent 1 hour. The sequence takes the minimum concurrent number of the cache stability server in the most recent 1 hour as an element of the set KPT. Use KPT(i) to represent the minimum concurrent number of the cache stability server with serial number i in the most recent 1 hour; set an empty sequence as the maximum concurrent number set KPU of the cache stability server in the most recent 1 hour. The sequence takes the maximum concurrent number of the cache stability server in the most recent 1 hour as an element of the set KPU. Use KPU(i) to represent the maximum concurrent number of the cache stability server with serial number i in the most recent 1 hour; obtain the average concurrent number in KPU and denote it as KPUs.

[0065] Among them, the minimum concurrent number of the cache stability server in the most recent 1 hour represents the lowest value of the concurrent request numbers received by the cache stability server in the past 1 hour. The maximum concurrent number of the cache stability server in the most recent 1 hour represents the highest value of the concurrent request numbers received by the cache stability server in the past 1 hour.

[0066] Among them, the concurrent response time difference includes the first concurrent response time difference and the second concurrent response time difference. Take the difference between STSx and STSm as the first concurrent response time difference SE; take the difference between STSm and STSs as the second concurrent response time difference SR;

[0067] S302, calculate the concurrent load request duration of the cache stability server through the concurrent response time difference of the cache stability server;

[0068] Furthermore, set an empty sequence POT to store the concurrent load request duration of the cache stability server. The sequence POT takes the concurrent load request duration POT(i) of the cache stability server as an element of the sequence POT. The calculation method of the concurrent load request duration POT(i) is as follows:

[0069] POT(i) = STS(i) + |(SE - SR)| × (KPT(i) + KPU(i)) / KPUs;

[0070] Among them, || is the absolute value symbol, and the principle of the concurrent load request duration POT(i) is as follows: Based on the server's own response time STS(i), the fluctuation of the server's response time is considered through the concurrent response time difference (SE - SR). If SE is much larger than SR or much smaller than SR, it indicates that the maximum response time is far from the average value, while the minimum response time is close to the average value, meaning that the server's performance drops significantly under high load. If SE is close to SR, it indicates that the server's performance is relatively stable under both high load and low load, and it is suitable as the target server. The server's concurrent load capacity is considered through (KPT(i) + KPU(i)) / KPUs. If the concurrent request numbers KPT(i) and KPU(i) of a certain server are both high, it means that the server is under great pressure. If the server's concurrent load is low, the POT(i) value is small, meaning that the server can give priority to processing new requests.

[0071] S303. Determine the target server according to the concurrent load request duration of the cache - stable server.

[0072] Specifically, sort the cache - stable servers from small to large according to the size of the concurrent load request duration POT(i). The sequence number corresponding to the largest value in the load duration POT(i) is G, and the sequence number corresponding to the smallest value in the load duration POT(i) is 1. Preferentially select the cache - stable server with a small sequence number as the target server according to the sequence numbers of the cache - stable servers.

[0073] S400. Generate and send a personalized order - following notice to the target user through the target server to send an operation instruction for order - making decisions.

[0074] Specifically, in step S400, the target server analyzes the potential needs of the user based on the user's historical booking behavior, browsing records, preference settings, and real - time market dynamics, and generates targeted order - following notices, such as price - change reminders, housing - shortage warnings, and time - limited discount pushes. Further, the target server sends personalized order - following notices to the target user through multiple channels (such as text messages, emails, or APP pushes, etc.), prompting the user to make an order decision as soon as possible, thereby improving the booking conversion rate and optimizing the user experience.

[0075] It should be noted that the logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in combination with an instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other suitable processing as necessary, and then stored in a computer memory.

[0076] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having suitable combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0077] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0078] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. are based on the orientation or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention.

[0079] In addition, the terms "first", "second", etc. used in the embodiments of the present invention are only for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated in this embodiment. Thus, the features defined with the terms "first", "second", etc. in the embodiments of the present invention may explicitly or implicitly indicate that at least one such feature is included in this embodiment. In the description of the present invention, the meaning of the word "plural" is at least two or more than two, such as two, three, four, etc., unless otherwise explicitly and specifically defined in the embodiment.

[0080] In the present invention, unless otherwise explicitly specified or limited in the embodiments, the terms "mounted", "connected", "coupled" and "fixed" and the like appearing in the embodiments should be understood in a broad sense. For example, the connection can be a fixed connection, a detachable connection, or integrated. It can be understood that it can also be a mechanical connection, an electrical connection, etc.; of course, it can also be directly connected, or indirectly connected through an intermediate medium, or it can be the communication inside two elements, or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific implementation circumstances.

[0081] In the present invention, unless otherwise explicitly specified and limited, the first feature being "on" or "under" the second feature may be that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, the first feature being "above", "over" and "on top of" the second feature may be that the first feature is directly above or obliquely above the second feature, or merely indicates that the first feature has a higher horizontal height than the second feature. The first feature being "under", "below" and "beneath" the second feature may be that the first feature is directly below or obliquely below the second feature, or merely indicates that the first feature has a lower horizontal height than the second feature.

[0082] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for realizing intelligent follow-up of hotel distribution based on enterprise private domain knowledge, characterized in that: The method comprises the following steps: S100, obtaining server information of the hotel; S200, performing dynamic cache frequency-effect analysis on the server based on the server information to determine a cache-stable server; S300, obtaining a target server by calculating a concurrent response time difference of a cache stable server; S400, generating and sending a personalized order notification to a target user through a target server to send an operation instruction for order decision.

2. According to claim 1, a method for realizing intelligent follow-up of hotel distribution based on enterprise private domain knowledge is characterized in that: In step S100, the server information of the hotel is obtained, wherein the hotel server information includes the response time of the server, the number of requests of the server, the number of concurrent requests of the server and the cache hit rate.

3. According to claim 1, a method for realizing intelligent follow-up of hotel distribution based on enterprise private domain knowledge is characterized in that: Let j represent the ID of each server, j = 1, 2, ..., K, where K represents the number of servers. Let r(j) be the maximum cache hit rate of the jth server in the last hour, p(j) be the minimum cache hit rate of the jth server in the last hour, and Ms be the average minimum cache hit rate of all servers in the last hour. Performing dynamic cache frequency effect analysis on the server, wherein the steps of performing dynamic cache frequency effect analysis on the server include: S201, calculating the cache loss of the server in the last hour; Calculate the j-th server cache loss ZG(j) in the last hour, where ZG(j) is the difference between r(j) and p(j). Add all ZG(j) and divide by K to get the stable cache loss HDC. S202, define an integer variable k, set the initial value to 1, and create a blank sequence ZRE; S203, performing dynamic cache frequency effect analysis, wherein the dynamic cache frequency effect analysis is: comparing the values ​​of ZG(j) and HDC, and comparing the values ​​of [p(j)+HDC] and Ms, if ZG(j) is less than HDC and [p(j)-HDC] is greater than Ms, then adding ZG(j) to the sequence ZRE; S204, judging whether the dynamic cache frequency effect analysis is completed, the specific judging method is: if the current variable k is less than K, then k is increased by 1, and the process returns to step S203 to continue the dynamic cache frequency effect analysis; if the current variable k is equal to K, then it means that the dynamic cache frequency effect analysis has been processed and the process goes to step S205; S205, record the servers corresponding to all elements in the sequence ZRE as cache stable servers.

4. According to claim 1, a method for realizing intelligent follow-up of hotel distribution based on enterprise private domain knowledge is characterized in that: In step S300, obtaining the target server by calculating the concurrent response time difference of the cache stable server includes the following steps: S301, calculating the concurrent response time difference of the cache stable server; S302, calculating the concurrent load request duration of the cache stable server by the concurrent response time difference of the cache stable server; S303, determining a target server according to the concurrent load request duration of the cache stable server.

5. According to claim 4, a method for realizing intelligent follow-up of hotel distribution based on enterprise private domain knowledge is characterized in that: In step S301, calculating the concurrent response time difference of the cache stabilization server includes: Set an empty sequence STS as the response time set of the cache stable server, where the sequence takes the response time of the cache stable server as the element of the set STS, and use STS(i) to represent the average response time of the cache stable server with sequence number i in the last hour, where i represents the sequence number of the cache stable server, i = 1, 2, ..., G, G is the number of cache stable servers; get the shortest response time in STS and record it as STSs; get the longest response time in STS and record it as STSx; get the average value of all response times in STS as STSm, set an empty sequence as the cache stable service The minimum concurrent number set KPT of the cache stable server in the last hour, wherein the sequence takes the minimum concurrent number of the cache stable server in the last hour as the element of the set KPT, and uses KPT(i) to represent the minimum concurrent number of the cache stable server in the last hour with the serial number i; an empty sequence is set as the maximum concurrent number set KPU of the cache stable server in the last hour, wherein the sequence takes the maximum concurrent number of the cache stable server in the last hour as the element of the set KPU, and uses KPU(i) to represent the maximum concurrent number of the cache stable server in the last hour with the serial number i; the average number of requests in KPU is obtained and recorded as KPUs; Among them, the minimum concurrent number of the cache stable server in the last hour indicates the lowest value of the concurrent requests received by the cache stable server in the past hour; the maximum concurrent number of the cache stable server in the last hour indicates the highest value of the concurrent requests received by the cache stable server in the past hour; The concurrent response time difference includes a first concurrent response time difference and a second concurrent response time difference. The difference between STSx and STSm is taken as the first concurrent response time difference SE; the difference between STSm and STSs is taken as the second concurrent response time difference SR.

6. A method for realizing intelligent follow-up of hotel distribution based on enterprise private domain knowledge according to claim 5, characterized in that: In step S302, calculating the concurrent load request duration of the cache stable server through the concurrent response time difference of the cache stable server includes: An empty sequence POT is set to store the concurrent load request duration of the cache stable server, wherein the sequence POT uses the concurrent load request duration POT(i) of the cache stable server as an element of the sequence POT.

7. A method for realizing intelligent follow-up of hotel distribution based on enterprise private domain knowledge according to claim 6, characterized in that: In step S303, determining the target server according to the concurrent load request duration of the cache stable server includes: According to the concurrent load request duration POT(i), the cache stable servers are sorted from small to large. The sequence number corresponding to the largest value in the load duration POT(i) is G, and the sequence number corresponding to the smallest value in the load duration POT(i) is 1. According to the sequence number of the cache stable server, the cache stable server with a small sequence number is preferentially selected as the target server.

8. According to claim 1, a method for realizing intelligent follow-up of hotel distribution based on enterprise private domain knowledge is characterized in that: In step S400, a personalized follow-up notification is generated and sent to the target user through the target server to send an operation instruction of order decision, including: the target server analyzes the user's potential needs based on the user's historical booking behavior, browsing history, preference settings and real-time market dynamics, and generates targeted follow-up notifications, such as price change reminders, housing shortage warnings, and limited-time discount push.

Images