Big data type search system based on 5G communication technology and search method thereof

Abstract

The invention discloses a big data type search system based on the 5G communication technology and a search method thereof, and relates to the technical field of internet search, the big data type search system comprises a data acquisition module, a data processing module, an analysis module and an application module, the priority of an ith query, the score of a pth query result and the response time of the query are calculated through the analysis module, and the query response time is calculated through the application module; according to the method, the query priority is dynamically adjusted through multiple dimensions such as correlation, data volume, geographic position, time delay and the like, the urgent query with high correlation is preferentially processed, the problem of inaccurate priority ranking of a traditional system is solved, the system flexibly adjusts the query sequence and the strategy to improve the efficiency, the user waiting time is shortened, and the user experience is improved. The query results are intelligently sorted according to the priorities, the click rates and the user scores, it is ensured that the most relevant results are preferentially presented, meanwhile, by analyzing the query response time and dynamically optimizing the processing flow, the response time is further shortened, and the user experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of Internet search technology, and specifically to a big data search system and a search method thereof based on 5G communication technology. Background Art

[0002] With the rapid development and application of 5G communication technology, the ability to collect, transmit and process big data has been significantly improved, especially in the field of large-scale information processing and search systems. 5G technology, with its extremely low latency, high bandwidth and powerful data transmission capabilities, has greatly promoted the development of cloud computing, big data and artificial intelligence, enabling these technologies to be better applied and improved in terms of real-time performance and data processing capabilities.

[0003] However, the response time of the existing search system in actual application may be affected by many factors. Under high load, the query response time of the system may be unstable and delayed, which in turn limits the low-latency advantage of 5G. Therefore, the real-time performance of the query process may not be fully guaranteed. The existing search system may adopt a fixed processing order when querying, and may not be able to dynamically adjust the query processing priority according to the real-time needs and urgency of the query. Therefore, high-priority queries may not be processed first, affecting the user experience and system efficiency. In addition, when facing large-scale data queries, the existing system sorting mechanism may not be able to accurately optimize the sorting of query results. The sorting of query results usually needs to consider multiple factors such as user evaluation, click status, and relevance. Therefore, the existing search system often has problems such as inaccurate result sorting and slow response speed, and cannot efficiently process all query requests. Summary of the Invention

[0004] The purpose of the present invention is to provide a big data search system and a search method thereof based on 5G communication technology, which solves the problems raised in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention provides the following big data search system based on 5G communication technology, comprising:

[0006] Data collection module: The data collection module collects network connection delay data, query data size information and query type information;

[0007] Data processing module: The connection delay data, query data size information, and query type information are input into the data processing module. The data processing module normalizes the input data to output the connection delay of the i-th query, the i-th query data size, and the time sensitivity coefficient of the i-th query;

[0008] Analysis module: The connection delay, data size, and time sensitivity coefficient of the i-th query are input into the analysis module. The analysis module outputs the priority of the i-th query, the score of the p-th query result, and the query response time.

[0009] Application module: The priority of the i-th query, the score of the p-th query result, and the query response time are input into the application module. The application module adjusts the query order based on the priority of the i-th query. The application module optimizes the display order of the query results based on the score of the p-th query result. The application module optimizes the system's resource scheduling and load balancing based on the query response time.

[0010] Optionally, the analysis module includes a query submodule, a query result sorting submodule and a response submodule.

[0011] Optionally, the calculation formula of the query submodule is as follows:

[0012]

[0013] in:

[0014] QPL i Refers to the priority of the i-th query, QA refers to the weight coefficient of the query factor, QB refers to the weight coefficient of the data size, and QC refers to QPLD i The weight coefficient, QD refers to QPLE i Weight coefficient, QPLA i Refers to the relevance score of the i-th query, QPLB i Refers to the connection delay of the i-th query, QPLC i Refers to the data size of the i-th query, QPLD i Refers to the geographic location correlation coefficient of the i-th query, QPLE i Refers to the time sensitivity coefficient of the i-th query;

[0015] QPLA i ×QPLB i Refers to the relationship between the urgency of the query and the actual join condition;

[0016] Refers to the priority of queries when processing large amounts of data;

[0017] QC×QPLD i +QD×QPLE i Refers to the combined impact of geographic location correlation and time sensitivity coefficient.

[0018] The processing process of the query submodule is as follows: set the connection delay QPLB of the i-th query i、Data size of the i-th query QPLC i and the time sensitivity coefficient QPLE of the i-th query i Input to the query submodule, the query submodule outputs the priority QPL of the i-th query i .

[0019] Optionally, the calculation formula of the query result sorting submodule is as follows:

[0020] RLD p =QPL i ×RD+QPLA i ×RA+QPLB p ×RB+QPLC p ×RC;

[0021] in:

[0022] RLD p Refers to the score of the p-th result, RA refers to the weight factor of the relevance, RLDB p Refers to the click value of the p-th result, RLDC p Refers to the user rating of the p-th result, RB refers to the weight factor of the click-through rate, RC refers to the weight factor of the user rating, and RD refers to the weight factor of the query priority;

[0023] The processing process of the query result sorting submodule is as follows: the priority QPL of the i-th query i and the relevance score QPLA of the i-th query i Input to the query result sorting submodule, the query result sorting submodule outputs the score RLD of the pth result p , and then sort all the query results, with the query results with high scores placed in front to be displayed to the user.

[0024] Optionally, the calculation formula of the response submodule is as follows:

[0025]

[0026] in:

[0027] TTP refers to the query response time, TA refers to RLD p TB refers to the weight coefficient of TTPA, TC refers to the weight coefficient of system factors, and TD refers to QPL i The weight coefficient, TTPA refers to the number of servers, TTPB refers to the load value of the system, TTPC refers to the network delay,

[0028] Refers to the relationship between each sorting result and the number of servers;

[0029] Refers to the combined impact of system load and network latency;

[0030] The processing process of the response submodule is as follows: the priority QPL of the i-th query i and the score of the p-th result RLD p Input to the response submodule, and the response submodule outputs the query response time TTP.

[0031] Optionally, the relevance score QPLA of the i-th query in the query submodule i The calculation formula is:

[0032]

[0033] Among them: TF j Refers to the frequency of the jth keyword in the query, IDF j refers to the inverse document frequency of the jth keyword in all documents, and m refers to the total number of keywords.

[0034] Optionally, the data processing module first cleans and then normalizes the input data. Data cleaning is used to remove invalid, duplicate and erroneous data.

[0035] The present invention also provides a search method for a big data search system based on 5G communication technology, comprising the following steps:

[0036] Step I: Collect network connection delay data, query data size information and query type information through the data acquisition module

[0037] Step II: Input the connection delay data, query data size information, and query type information into the data processing module. The data processing module normalizes the input data and outputs the connection delay of the i-th query, the i-th query data size, and the time sensitivity coefficient of the i-th query;

[0038] Step III: Input the connection delay, data size, and time sensitivity coefficient of the ith query into the analysis module, and output the priority of the ith query, the score of the pth query result, and the query response time.

[0039] Step IV: Input the priority of the i-th query, the score of the p-th query result, and the query response time into the application module. The application module adjusts the query order based on the priority of the i-th query. The application module optimizes the display order of the query results based on the score of the p-th query result. The application module optimizes the system's resource scheduling and load balancing based on the query response time.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] 1. The present invention outputs the priority of the i-th query through the query submodule. The query submodule can accurately calculate the priority for each query, so that the system can quickly determine which queries need to be processed first. In the low-latency environment of the 5G network, the system can quickly determine the importance of the query and give priority to high-priority queries. This can ensure high-real-time queries and avoid wasting precious bandwidth and computing resources for low-priority queries. The low-latency characteristics of 5G technology combined with the query correlation and connection delay in the query submodule greatly shorten the delay in the query processing process. When users are widely geographically distributed, the introduction of geographical location correlation and time sensitivity in the query submodule can ensure that users' queries in different regions and at different times can receive timely responses.

[0042] 2. The present invention outputs the score of the pth result through the query result sorting submodule. The query result sorting submodule sorts the query results so that the most relevant and important results are returned to the user first. The low latency and high-speed transmission under the 5G network ensure that users can quickly obtain real-time search results. The weighted multiplication formula makes the search results more in line with user expectations and interests by integrating relevance, click-through rate and user ratings. 5G technology enables the system to receive user behavior data in real time to ensure that users obtain a more accurate and targeted search experience. The search results obtained by users are more relevant to reduce the user's waiting time and improve the response speed and accuracy of the search system. Based on the fast network and low latency characteristics of 5G and combined with the query result sorting submodule, the ranking of search results can be updated in real time. In high-concurrency scenarios, the system can quickly and accurately return the most relevant content to the user. The query result sorting submodule can ensure that real-time, high-priority query results are displayed in the shortest time.

[0043] 3. The present invention outputs the response time of the query through the response submodule. The function of the response submodule is to optimize the real-time response capability of the system. Under high load, the system can balance the load and delay to ensure a quick response to the query. Under high load, the system can balance the load and delay to ensure a quick response to the query. Based on the 5G environment, the response submodule ensures that the system can maintain a low-latency response under high concurrency. By dynamically adjusting server resources, such as the number of servers and system load, the response submodule can effectively avoid response delays caused by server overload. In the case of a surge in user queries, the low-latency characteristics of 5G communication can ensure that a large number of query requests can be processed in a very short time under the influence of delay and query complexity. By adjusting the relationship between delay and server resource allocation based on the response submodule, it is ensured that the system can still provide stable services under high load. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flowchart of the method steps of the big data search system based on 5G communication technology;

[0045] Figure 2 This is a schematic diagram of the overall structure of the big data search system based on 5G communication technology;

[0046] Figure 3 This is a structural diagram of the analysis module in the big data search system based on 5G communication technology. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] Regarding this big data search system based on 5G communication technology, it is different from existing search systems. The response time of existing search systems in actual big data environments is still affected by multiple factors. Especially under high load conditions, the system's query response time may still be unstable and have high latency, which limits the advantage of 5G's low latency, and may result in the real-time performance of the query process not being fully guaranteed. Existing search systems may adopt a fixed processing order when processing queries, and may not be able to dynamically adjust the query processing priority according to the real-time needs and urgency of the query, which may result in high-priority queries not being processed first, thereby affecting user experience and system efficiency. In addition, in the face of large-scale data queries in a big data environment, the sorting mechanism of existing systems often cannot accurately optimize the sorting of query results. The sorting of query results usually needs to consider multiple factors such as user evaluation, click status, and relevance. As a result, existing search systems often have problems such as inaccurate result sorting and slow response speed, and cannot efficiently process all query requests.

[0049] The search system calculates the query priority and comprehensively considers factors such as the relevance, data volume, geographic location, and latency of the query. It can adjust the query priority in real time. For urgent and highly relevant queries, the system can give priority to them, avoiding the problem of inaccurate priority sorting in traditional search systems. The system can flexibly adjust the order and strategy of query processing according to the actual needs of different queries, improve query processing efficiency, and reduce user waiting time. The query results can be intelligently sorted according to the query priority, click-through rate, and user rating, so as to ensure that the most relevant and urgent query results can be presented first, avoiding users from wasting too much time on irrelevant information in the search results, and optimizing the search efficiency in large-scale data environments. In addition, by introducing the analysis of the query response time, the system can dynamically adjust the query processing process in real time according to factors such as the priority of the query request, network latency, and system load, thereby reducing the query response time. The present invention can automatically adjust the query strategy to ensure that the query request can be responded to in the shortest time, greatly improving the user experience and system efficiency.

[0050] Example 1: Please refer to Figures 1 to 3 This implementation provides a big data search system based on 5G communication technology, including:

[0051] Data collection module: The data collection module collects network connection delay data, query data size information and query type information;

[0052] Data processing module: The connection delay data, query data size information, and query type information are input into the data processing module. The data processing module normalizes the input data to output the connection delay of the i-th query, the i-th query data size, and the time sensitivity coefficient of the i-th query;

[0053] Analysis module: The connection delay, data size, and time sensitivity coefficient of the i-th query are input into the analysis module. The analysis module outputs the priority of the i-th query, the score of the p-th query result, and the query response time.

[0054] Application module: The priority of the i-th query, the score of the p-th query result, and the query response time are input into the application module. The application module adjusts the query order based on the priority of the i-th query, optimizes the display order of the query results based on the score of the p-th query result, and optimizes the system's resource scheduling and load balancing based on the query response time.

[0055] The analysis module includes a query submodule, a query result sorting submodule and a response submodule.

[0056] In this embodiment: Based on the query submodule, the priority of each query can be accurately calculated, so that the system can quickly determine which queries need to be processed first. This is crucial for big data search systems based on 5G communication technology. In the low-latency environment of the 5G network, the system can quickly determine the importance of the query and give priority to high-priority queries. This can ensure that high-real-time queries, such as real-time traffic, stock quotes, and video streaming queries, are responded to first, avoiding the waste of precious bandwidth and computing resources for low-priority queries. The low-latency characteristics of 5G technology combined with the query relevance and connection delay in the query submodule greatly shorten the delay in the query processing process. Especially when users are geographically distributed widely, the introduction of geographic location relevance and time sensitivity in the query submodule can ensure that users' queries in different regions and at different times can receive timely responses.

[0057] The query result sorting submodule prioritizes query results so that the most relevant and important results are returned to the user first. By combining the query priorities in the query submodules, the ranking of each search result can be dynamically adjusted, thereby optimizing the user's search experience. High-priority query results, such as real-time news and streaming content, are displayed first, while complex queries are ranked later. This can improve the system's response efficiency, especially in a big data environment where the system needs to find the most relevant information from massive amounts of data. The query result sorting submodule reduces the interference of irrelevant information in the search results through weighted sorting. Based on the fast network and low latency characteristics of 5G, combined with the query result sorting submodule, the ranking of search results can be updated in real time. Especially in high-concurrency scenarios, the system can quickly and accurately return the most relevant content to the user. For example, in an intelligent transportation system, a user's query may involve real-time traffic data. The query result sorting submodule can ensure that real-time, high-priority query results are displayed in the shortest possible time.

[0058] The role of the response submodule is to optimize the real-time response capability of the system, ensuring that the system can balance the load and delay under high load, and ensure fast response to queries. In a 5G-based environment, the response submodule ensures that the system can maintain low-latency response under high concurrency. By dynamically adjusting server resources, such as the number of servers and system load, the response submodule can effectively avoid response delays caused by server overload, especially when the number of user queries surges. The low-latency characteristics of 5G communication can ensure that a large number of query requests can be processed in a very short time despite the influence of delay and query complexity. Based on the response submodule, by adjusting the relationship between delay and server resource allocation, the system can ensure that it can still provide stable services under high load.

[0059] See also Figures 1 to 3 , the processing process of the query submodule is as follows:

[0060]

[0061] in:

[0062] QPL i Refers to the priority of the i-th query, which determines the order in which queries are processed in the entire system. Queries with higher priorities will be processed first to ensure timely responses;

[0063] QA refers to the weight coefficient of the query factor, QB refers to the weight coefficient of the data size, and QC refers to QPLD i The weight coefficient, QD refers to QPLE i The weight coefficient of

[0064] QPLA i Refers to the relevance score of the i-th query, using TF-IDF term frequency-inverse document frequency to calculate the frequency of occurrence of the query keyword in the document. Relevance is usually calculated by matching the query keyword with the data in the database and then needs to be normalized;

[0065]

[0066] Among them: TF j Refers to the frequency of the jth keyword in the query, IDF j refers to the inverse document frequency of the jth keyword in all documents, and m refers to the total number of keywords;

[0067] QPLB i The connection latency of the i-th query represents the network delay for the query request to reach the server. Latency is typically very low on 5G networks, but the real-time nature of the network still needs to be considered. This can be monitored in real time using network latency measurement tools. Use the Ping command and Traceroute tools to measure the latency from the query source to the server, then normalize this value and enter it into this formula: divide the connection latency of the i-th query by the maximum query connection latency.

[0068] QPLC iRefers to the data size of the i-th query, indicating the size of the data block stored in the database. Larger data sets require more time and computing resources to process, and are not directly accessible through database or storage system management tools. Typically, data storage systems, such as SQL databases and NoSQL databases, can provide statistical information about the size of data tables or data sets and use normalization processing methods. Normalization processing methods are well-known technical means. In this system, the data size of the i-th query needs to be divided by the maximum size of the data to be retrieved before entering it into this formula for normalization processing;

[0069] QPLD i Refers to the geographic location correlation coefficient of the i-th query, indicating the degree of geographic location matching, which usually ranges from 0 to 1. If the query location and the user location are very close, the correlation is 1, and if the distance is far, it is close to 0;

[0070] QPLE i Refers to the time sensitivity coefficient of the i-th query. The time sensitivity of the query ranges from 0 to 1. Time sensitivity is usually determined by the query type. For example, real-time queries have higher time sensitivity, while non-real-time queries have lower sensitivity. For example, real-time queries such as stocks and news are assigned higher time sensitivity, while historical queries such as long-term weather forecasts are assigned lower time sensitivity.

[0071] QPLA i ×QPLB i Refers to the relationship between the urgency of the query and the actual connection conditions. If the query relevance is high, that is, the query is more important to the user, but the connection latency is long, this will increase the query processing burden and require more resources to process. Low latency is a key feature in 5G systems. Therefore, optimizing the combination of connection latency and query relevance helps to quickly return relevant information in low-latency situations, improving system response speed.

[0072] Refers to the priority of queries when processing large amounts of data. Queries with large data volumes require more time to process, so high data volumes affect query priority. The system needs to be optimized to avoid processing delays caused by excessive data volumes. In the 5G environment, data throughput has increased significantly. The system needs to optimize this part to balance data transmission and computing time to ensure fast response times even with large data sets.

[0073] QC×QPLD i +QD×QPLE iRefers to the combined impact of geographic location relevance and time sensitivity coefficients. If geographic location relevance and time sensitivity are high, then this part will increase the query priority. 5G communication technology makes the acquisition of geographic location and time sensitivity more accurate. Therefore, it can prioritize the most relevant content based on the user's real-time location and needs, thereby improving the user experience.

[0074] The processing process of the query submodule is as follows: set the connection delay QPLB of the i-th query i 、Data size of the i-th query QPLC i and the time sensitivity coefficient QPLE of the i-th query i Input to the query submodule, the query submodule outputs the priority QPL of the i-th query i .

[0075] In this embodiment: the priority of the query is determined in this submodule, taking into account factors such as query relevance, connection delay, data size, geographic location and time sensitivity. Since the main purpose of this submodule is to optimize the priority of each query, the subscript i here represents different queries, and each query will have an independent set of parameters, such as query relevance, connection delay and data size. Therefore, we use the subscript i to distinguish the parameters of these queries. In this submodule, i refers to the query, and each query has independent parameters such as relevance, delay, data size, etc. The core goal of this submodule is to optimize the processing priority of the query based on the relevance, connection delay, data size, geographic location and time sensitivity of the query. By accurately calculating the priority for each query, the system can give priority to the most important or urgent queries to the user when resources are limited. For a big data search system based on 5G communication technology, it is crucial to reduce the processing time of irrelevant queries when processing a large number of queries. Since 5G communication technology provides extremely low latency and higher transmission rate, the connection delay and data size in the formula The impact can be reflected in the context of 5G networks, showing significant improvements, reducing query response time. At the same time, the weighted synthesis of geographic location relevance and time sensitivity further improves the accuracy and timeliness of responses in practical applications. For example, queries for real-time video content will have a higher priority than other ordinary queries, ensuring real-time performance. 5G technology provides high-speed transmission and low-latency support for big data queries. Each parameter in the formula, such as query relevance, latency, and geographic location relevance, can be quickly processed under the low-latency characteristics of the 5G network, optimizing the system's processing of large-scale data sets. The distributed processing capabilities of big data technology combined with the low latency and high-speed bandwidth of 5G provide strong performance support for big data search systems. By optimizing the query priority calculation, Formula 1 can effectively utilize its low latency advantage under the 5G network, improving the ability to quickly locate and accurately query in massive data. In big data scenarios, massive query requests and massive data storage require efficient sorting and priority processing. Formula 1 optimizes this data flow and improves the efficiency of big data processing systems.

[0076] See also Figures 1 to 3 , the processing process of the query result sorting submodule is as follows:

[0077] RLD p =QPL i ×RD+QPLA i ×RA+QPLB p ×RB+QPLC p ×RC;

[0078] in:

[0079] RLD prefers to the score of the p-th result, and RA refers to the weight factor of the correlation;

[0080] RLDB p Refers to the click-through value of the p-th result. The click-through value, or click-through rate, is the ratio of the number of users who clicked on a search result to the total number of users who displayed the result.

[0081] RLDC p Refers to the user rating of the p-th result. The user rating can be based on the user's feedback on the result, in the form of a 10-point rating, and then normalized to a range of 0-1. The score given by the user to the search result is usually given on the page;

[0082] RB refers to the weight factor of click rate, RC refers to the weight factor of user rating, and RD refers to the weight factor of query priority;

[0083] The processing process of the query result sorting submodule is as follows: the priority QPL of the i-th query i and the relevance score QPLA of the i-th query i Input to the query result sorting submodule, the query result sorting submodule outputs the score RLD of the pth result p , and then sort all the query results, with the query results with high scores placed in front to be displayed to the user.

[0084] In this embodiment: the subscript p that appears in this submodule refers to the query result rather than the query itself. The purpose of this submodule is to sort the query results, so the subscript p here represents the sorted query result. Parameters such as the priority of each query result will affect the sorting. In a big data environment based on 5G communication technology, the amount of data is usually very large. This submodule avoids the item-by-item summation calculation of each query result by using the weighted multiplication method, and can quickly sort the data according to multiple important factors, thereby improving the processing efficiency of the system. Even if the amount of data is huge, the system can still maintain a high response speed and accuracy, ensuring that users can quickly Obtain query results that meet the needs. By adjusting the weights of RA, RB, RC, and RD, the priorities of relevance, click-through rate, and score can be adjusted according to different needs and different time periods. Dynamic adjustment of these parameters can better adapt to the preferences of different users and real-time changing environments, such as hot topics and real-time content, so that search results are always the latest and most relevant. The low latency and high-speed transmission under the 5G network ensure that users can quickly obtain real-time search results. The weighted multiplication formula makes the search results more in line with user expectations and interests by integrating relevance, click-through rate, and user score. 5G technology enables the system to receive user behavior data in real time, such as clicks and scores, so as to dynamically adjust the ranking of results, ensuring that users have a more accurate and targeted search experience. The search results users receive are more relevant, reducing user waiting time and improving the response speed and accuracy of the search system. Query results are not only based on static data, such as the relevance score at the time of query, but can be optimized based on real-time user interaction data. This enables the search system to maintain high responsiveness and accuracy when facing changing needs and content. The sorting mechanism of this submodule helps the system in application scenarios with high real-time requirements, such as real-time data monitoring or intelligent In the transportation system, users’ needs can be quickly responded to. In a big data environment, the amount of query results is usually huge. This module accurately controls the priority of result display through click-through rate, user rating and query relevance, enabling the system to efficiently sort and filter massive amounts of data, especially in 5G networks. The system can quickly process and return results. In the low-latency, high-bandwidth network environment of 5G, the sorting of search results is not only fast but also accurate. The powerful bandwidth and low latency of the 5G network support the rapid processing of large-scale data. Combined with the efficient computing of the big data processing framework, this sorting mechanism can process a large number of search requests and data in a short time.

[0085] See also Figures 1 to 3 , the processing of the response submodule is as follows:

[0086]

[0087] in:

[0088] TTP refers to the query response time, TA refers to RLD p TB refers to the weight coefficient of TTPA, TC refers to the weight coefficient of system factors, and TD refers to QPL i The weight coefficient of

[0089] TTPA refers to the number of servers. Use data center management tools such as Kubernetes or server monitoring tools to obtain the current number of servers. Then use normalization to normalize the number of servers and input it into this formula. That is, divide the number of servers by the maximum number of servers.

[0090] TTPB refers to the system load value. System load can be monitored through indicators such as CPU usage, memory usage, or database load. It is usually obtained through operating system monitoring tools such as htop or top command. Usually, the value is expressed as a percentage and then divided by 100 to make it a value between 0 and 1;

[0091] TTPC refers to network latency. The current network latency is obtained using the Ping command, Traceroute tool, or 5G network monitoring tool. This value is then normalized and input into this formula. The specific normalization method is common in existing technologies: divide the network latency by the maximum value of the network latency.

[0092] Refers to the relationship between each ranking result and the number of servers. The ranking of query results needs to be distributed and processed by multiple servers. The number of servers will affect the response time of the query. If there are more servers, the burden on each server can be reduced, thereby improving the response speed of the system.

[0093] The weight coefficient TC refers to the combined impact of system load and network latency. It is used to balance the effects of load and latency. If the system load is high and the latency is low, the response time may be relatively slow. However, if the load is too high, the response time may increase significantly. The low latency of 5G makes the impact of network latency on query response time relatively small, while the system load becomes more important. By optimizing load distribution and latency control, we can ensure that big data queries can achieve real-time responses in the 5G environment.

[0094] The processing process of the response submodule is as follows: the priority QPL of the i-th query i and the score of the p-th result RLD p Input to the response submodule, and the response submodule outputs the query response time TTP.

[0095] In this embodiment: This submodule outputs the query response time TTP by considering factors such as server load, number of results, and network latency, providing a basis for helping managers optimize query response time. The function of this submodule is to optimize the response time of real-time queries based on the current system load, number of servers, and query response latency. For big data systems based on 5G communication, this submodule dynamically adjusts network load and server resource allocation to make query responses faster. Especially when the system load is too high, this submodule can help allocate resources reasonably to ensure that the query response time is completed in the shortest time. The low latency characteristics of the 5G network are particularly important for real-time response. Big data systems often need to be dynamically adjusted according to system load and latency. This submodule is designed to serve this purpose. Through flexible delay control and load adjustment, the system response time is optimized to ensure that users can obtain query results in the shortest time when processing big data. The low latency and high bandwidth of 5G technology can significantly improve the effect of real-time response optimization in this submodule, especially under high load conditions, ensuring that system parameters can be quickly adjusted to reduce response delays.

[0096] It is worth noting that the query response time TTP is further calculated to affect the weight coefficient QA of the query factor in the query submodule, and then the priority QPL of the i-th query i , the score of the p-th result RLD p The query response time TTP is continuously optimized. The specific processing process is as follows:

[0097] First: QA new =QA old -ES×[(QPLB i ×QPLA i ×TTP) / QPLC i ];

[0098] Second: Set the iteration termination condition:

[0099] Termination condition 1: The number of iterations is 100;

[0100] Termination condition 2: |QA new -QA old |<0.001;

[0101] in:

[0102] QA new Refers to the weight coefficient of the query factor after iteration, QA old Refers to the weight coefficient of the query factor before iteration, ES refers to the learning rate, and controls the iteration step size.

[0103] In this embodiment, by optimizing the weight coefficient QA of the query factor, the connection delay QPLB of the i-th query is optimized and influenced. i The purpose is to accurately control the impact of connection delay on query priority, so that the priority calculation is more in line with the actual system load, network conditions and query urgency. Since the iterative adjustment of QA can reflect the system status, such as load and network conditions, in real time, the query response speed can be effectively optimized, especially in 5G networks. Low latency makes the iterative process more efficient. Through iterative optimization of weighted QA, the influence of connection delay can be adaptively adjusted in different query scenarios, thereby improving the query processing capability and user experience of the overall system. By introducing the weight coefficient QA, its role in query priority can be flexibly controlled, which enables the system to dynamically adapt to different query requirements and network conditions, and through optimizing and influencing the connection delay QPLB i The ability to significantly reduce query response time can fully utilize the low latency advantage of 5G networks, thereby improving the efficiency of big data query systems. In a big data environment, system load may fluctuate over time. By iteratively updating connection delays, the system's query processing strategy can be adjusted according to real-time load conditions, enabling the system to dynamically adapt to load changes, maintain good query performance, and reduce performance fluctuations in the system due to network delays or unstable connections, thereby ensuring that the system can maintain efficient and stable operation in different network environments.

[0104] In the specific implementation process, the multiple sub-modules in this method are used to form a big data search system. i 、Data size of the i-th query QPLC i and the time sensitivity coefficient QPLE of the i-th query i Input to the query submodule, the query submodule outputs the priority QPL of the i-th query i The query submodule can accurately calculate the priority for each query, so that the system can quickly determine which queries need to be processed first, which is crucial for big data search systems based on 5G communication technology. In the low-latency environment of the 5G network, the system can quickly determine the importance of the query and give priority to high-priority queries. This ensures that high-real-time queries, such as real-time traffic, stock quotes, and video streaming queries, are responded to first, avoiding low-priority queries from wasting precious bandwidth and computing resources. The low-latency characteristics of 5G technology combined with the query relevance and connection delay in the query submodule greatly shorten the delay in the query processing process. Especially when users are geographically distributed widely, the introduction of geographic location relevance and time sensitivity in the query submodule can ensure that users' queries in different regions and at different times can receive timely responses;

[0105] By setting the priority QPL of the i-th query i and the relevance score QPLA of the i-th query i Input to the query result sorting submodule, the query result sorting submodule outputs the score RLD of the pth result p Then, all query results are sorted, with query results with high scores placed first to be displayed to users. The query result sorting submodule prioritizes the query results, so that the most relevant and important results are returned to users first. The low latency and high-speed transmission under the 5G network ensure that users can quickly obtain real-time search results. The weighted multiplication formula makes the search results more in line with user expectations and interests by integrating relevance, click-through rate, and user ratings. 5G technology enables the system to receive user behavior data in real time, such as clicks and ratings, to dynamically adjust the ranking of results, ensuring that users have a more accurate and targeted search experience. Users receive more relevant search results, reducing user waiting time and improving the response speed and accuracy of the search system. The query result sorting submodule reduces the interference of irrelevant information in the search results through weighted sorting. Based on the fast network and low latency characteristics of 5G, combined with the query result sorting submodule, the ranking of search results can be updated in real time. Especially in high-concurrency scenarios, the system can quickly and accurately return the most relevant content to users. For example, in intelligent transportation systems, user queries may involve real-time traffic data. The query result sorting submodule can ensure that real-time, high-priority query results are displayed in the shortest time.

[0106] By setting the priority QPL of the i-th query i and the score of the p-th result RLD p Input to the response submodule, the response submodule outputs the query response time TTP. The function of the response submodule is to optimize the real-time response capability of the system, ensure that the system can balance the load and delay under high load, ensure fast response to queries, and ensure that the system can balance the load and delay under high load to ensure fast response to queries. In a 5G-based environment, the response submodule ensures that the system can still maintain low-latency response under high concurrency. By dynamically adjusting server resources, such as the number of servers and system load, the response submodule can effectively avoid response delays caused by server overload. In the case of a surge in user query volume, the low-latency characteristics of 5G communication can ensure that a large number of query requests can be processed in a very short time under the influence of delay and query complexity. By adjusting the relationship between delay and server resource allocation based on the response submodule, it is ensured that the system can still provide stable services under high load.

[0107] Example 2: Please refer to Figure 1 、 Figure 2 and Figure 3The data processing module first cleans and then normalizes the input data. Data cleaning is used to remove invalid, duplicate and erroneous data.

[0108] In this embodiment, data cleaning helps to pre-process the data, thereby ensuring efficient calculation of subsequent data and improving the accuracy of analysis and calculation.

[0109] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A big data search system based on 5G communication technology, characterized by: include: Data collection module: The data collection module collects network connection delay data, query data size information and query type information; Data processing module: The connection delay data, query data size information, and query type information are input into the data processing module. The data processing module normalizes the input data to output the connection delay of the i-th query, the i-th query data size, and the time sensitivity coefficient of the i-th query; Analysis module: The connection delay, data size, and time sensitivity coefficient of the i-th query are input into the analysis module. The analysis module outputs the priority of the i-th query, the score of the p-th query result, and the query response time. Application module: The priority of the i-th query, the score of the p-th query result, and the query response time are input into the application module. The application module adjusts the query order based on the priority of the i-th query. The application module optimizes the display order of the query results based on the score of the p-th query result. The application module optimizes the system's resource scheduling and load balancing based on the query response time.

2. The big data search system based on 5G communication technology according to claim 1, characterized in that: The analysis module includes a query submodule, a query result sorting submodule and a response submodule.

3. The big data search system based on 5G communication technology according to claim 2, characterized in that: The calculation formula of the query submodule is as follows: in: QPL i Refers to the priority of the i-th query, QA refers to the weight coefficient of the query factor, QB refers to the weight coefficient of the data size, and QC refers to QPLD i The weight coefficient, QD refers to QPLE i Weight coefficient, QPLA i Refers to the relevance score of the i-th query, QPLB i Refers to the connection delay of the i-th query, QPLC i Refers to the data size of the i-th query, QPLD i Refers to the geographic location correlation coefficient of the i-th query, QPLE i Refers to the time sensitivity coefficient of the i-th query; QPLA i ×QPLB i Refers to the relationship between the urgency of the query and the actual join condition; Refers to the priority of queries when processing large amounts of data; QC×QPLD i +QD×QPLE i Refers to the combined impact of geographic location correlation and time sensitivity coefficient. The processing process of the query submodule is as follows: set the connection delay QPLB of the i-th query i 、Data size of the i-th query QPLC i and the time sensitivity coefficient QPLE of the i-th query i Input to the query submodule, the query submodule outputs the priority QPL of the i-th query i .

4. The big data search system based on 5G communication technology according to claim 3, characterized in that: The calculation formula of the query result sorting submodule is as follows: RLD p =QPL i ×RD+QPLA i ×RA+QPLB p ×RB+QPLC p ×RC; in: RLD p Refers to the score of the p-th result, RA refers to the weight factor of the relevance, RLDB p Refers to the click value of the p-th result, RLDC p Refers to the user rating of the p-th result, RB refers to the weight factor of the click-through rate, RC refers to the weight factor of the user rating, and RD refers to the weight factor of the query priority; The processing process of the query result sorting submodule is as follows: the priority QPL of the i-th query i and the relevance score QPLA of the i-th query i Input to the query result sorting submodule, the query result sorting submodule outputs the score RLD of the pth result p , and then sort all the query results, with the query results with high scores placed in front to be displayed to the user.

5. The big data search system based on 5G communication technology according to claim 4, characterized in that: The calculation formula of the response submodule is as follows: in: TTP refers to the query response time, TA refers to RLD p TB refers to the weight coefficient of TTPA, TC refers to the weight coefficient of system factors, and TD refers to QPL i The weight coefficient, TTPA refers to the number of servers, TTPB refers to the load value of the system, TTPC refers to the network delay, Refers to the relationship between each sorting result and the number of servers; Refers to the combined impact of system load and network latency; The processing process of the response submodule is as follows: the priority QPL of the i-th query i and the score of the p-th result RLD p Input to the response submodule, and the response submodule outputs the query response time TTP.

6. The big data search system based on 5G communication technology according to claim 3, characterized in that: The relevance score QPLA of the i-th query in the query submodule i The calculation formula is: Among them: TF j Refers to the frequency of the jth keyword in the query, IDF j refers to the inverse document frequency of the jth keyword in all documents, and m refers to the total number of keywords.

7. The big data search system based on 5G communication technology according to claim 1, characterized in that: The data processing module first cleans and then normalizes the input data. Data cleaning is used to remove invalid, duplicate and erroneous data.

8. A search method using the big data search system based on 5G communication technology according to claim 1, characterized in that: The following steps are involved: Step I: Collect network connection delay data, query data size information and query type information through the data acquisition module Step II: Input the connection delay data, query data size information, and query type information into the data processing module. The data processing module normalizes the input data and outputs the connection delay of the i-th query, the i-th query data size, and the time sensitivity coefficient of the i-th query; Step III: Input the connection delay, data size, and time sensitivity coefficient of the ith query into the analysis module, and output the priority of the ith query, the score of the pth query result, and the query response time. Step IV: Input the priority of the i-th query, the score of the p-th query result, and the query response time into the application module. The application module adjusts the query order based on the priority of the i-th query. The application module optimizes the display order of the query results based on the score of the p-th query result. The application module optimizes the system's resource scheduling and load balancing based on the query response time.

Images