Search engine based data processing method and device, electronic equipment and medium

Abstract

Embodiments of the present application provide a data processing method and device based on a search engine, electronic equipment and medium, belonging to the technical field of data processing. The method comprises: obtaining user information of a plurality of users based on a search engine, wherein the user information comprises inquiry information of the user and the occurrence time of the inquiry information; dividing the inquiry information according to a preset time rule and the occurrence time to obtain a plurality of inquiry sets; performing semantic space calculation on the inquiry sets to obtain a correction tendency value of the inquiry sets; screening the correction tendency value according to a preset drift value to obtain a target correction tendency value; and adjusting the search engine according to the target correction tendency value. Embodiments of the present application can selectively analyze the semantics in the inquiry sentences of the user within a preset time, improving the efficiency of data analysis.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a data processing method, apparatus, electronic device and medium based on a search engine. Background Technology

[0002] With the development of the internet, search engine-based data processing has gradually become an important research branch in the field of data processing. Among them, the most common search engine-based data processing method is to analyze the user's search results list, that is, to analyze the order in which users click on the search results list, the number of clicks, and current hot keyword recommendations. This requires tracking and analyzing a large amount of information, resulting in low data processing efficiency. To solve this problem, there is currently another method, "user corrective question behavior capture," which analyzes data by capturing the behavior of "users sending multiple semantically similar questions in a short period of time." Since users may be dissatisfied with the search results list, they may try to correct their input by changing the question with the same meaning. However, in the process of behavior capture, there are often problems such as unclear unit time periods and inability to determine the semantic information of multiple questions, resulting in huge amounts of data analysis and low data analysis efficiency. Summary of the Invention

[0003] The main objective of this application is to propose a data processing method, apparatus, electronic device, and medium based on a search engine, which can selectively analyze the semantics of user queries within a preset time, thereby improving the efficiency of data analysis.

[0004] To achieve the above objectives, a first aspect of this application proposes a data processing method based on a search engine, the method comprising:

[0005] User information of multiple users is obtained based on a search engine, wherein the user information includes user query information and the time when the query information occurred;

[0006] The query information is divided according to preset time rules and the occurrence time to obtain multiple query sets;

[0007] Semantic space calculations are performed on the query set to obtain the correction tendency value of the query set;

[0008] The correction tendency value is filtered according to the preset drift value to obtain the target correction tendency value;

[0009] The search engine is adjusted based on the target correction tendency value.

[0010] In some embodiments, the preset diffusion model includes a text front-end module, wherein the query information is divided according to preset time rules and the occurrence time to obtain multiple query sets, including:

[0011] The query information is sorted according to the occurrence time to obtain an occurrence time sequence;

[0012] The occurrence time sequence is divided according to the preset time rules to obtain multiple time intervals;

[0013] By statistically analyzing all the query information within the time interval, multiple query sets are obtained.

[0014] In some embodiments, the query set includes multiple semantic vectors; the step of performing semantic space calculations on the query set to obtain the correction tendency value of the query set includes:

[0015] Obtain multiple semantic vectors from the query set;

[0016] The semantic vector is calculated based on a preset distance function to obtain the distance parameter of the semantic vector;

[0017] The semantic space value is obtained by averaging the distance parameters according to the clustering analysis algorithm.

[0018] The correction tendency value of the query set is obtained based on the semantic space value and the number of semantic vectors.

[0019] In some embodiments, the step of filtering the correction tendency value according to a preset drift value to obtain a target correction tendency value includes:

[0020] A tendency curve is generated based on the corrective tendency value and the occurrence time;

[0021] The tendency curve is divided according to the preset drift value to obtain the correction threshold corresponding to the correction tendency value;

[0022] The target correction tendency value is obtained by filtering the correction tendency values ​​based on the correction threshold.

[0023] In some embodiments, the step of filtering the correction tendency values ​​based on the correction threshold to obtain the target correction tendency value includes:

[0024] Compare the correction threshold with the correction tendency value;

[0025] When the comparison result shows that the correction tendency value is greater than the correction threshold, the correction threshold is determined as the target tendency value.

[0026] In some embodiments, the preset drift value includes drift variance and drift mean; the step of dividing the tendency curve according to the preset drift value to obtain a correction threshold corresponding to the correction tendency value includes:

[0027] The tendency curve is calculated based on the average drift value;

[0028] The calculated tendency curve is discretized based on the drift variance to obtain the correction threshold corresponding to the correction tendency value.

[0029] In some embodiments, it also includes:

[0030] The correction tendency value is monitored based on preset monitoring rules to determine the mutation tendency value;

[0031] The search engine is updated based on the mutation tendency value.

[0032] To achieve the above objectives, a second aspect of this application provides a data processing apparatus based on a search engine, the apparatus comprising:

[0033] The information acquisition module is used to acquire user information of multiple users based on a search engine, wherein the user information includes user query information and the time when the query information occurred;

[0034] The semantic segmentation module is used to segment the query information according to a preset time rule and the occurrence time to obtain multiple query sets;

[0035] The spatial computing module is used to perform semantic space calculations on the query set to obtain the correction tendency value of the query set;

[0036] The target screening module is used to screen the correction tendency value according to the preset drift value to obtain the target correction tendency value;

[0037] An engine adjustment module is used to adjust the search engine based on the target correction tendency value.

[0038] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the search engine-based data processing method as described in the first aspect.

[0039] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the search engine-based data processing method as described in the first aspect.

[0040] The data processing method, apparatus, electronic device, and storage medium based on a search engine proposed in this application first obtain user information from multiple users through a search engine, and then divide the query information according to preset time rules and the occurrence time of the query information in the user information, thereby obtaining multiple query sets divided by time, realizing the division of query information within a preset time. Then, semantic space calculation is performed on the query sets to obtain the correction tendency value of each query set, and the correction tendency value is filtered according to a preset drift value, thereby selecting the target correction tendency value from multiple correction tendency values. Finally, the search engine is adjusted according to the target correction tendency value, thereby enabling selective optimization of the search engine and improving the efficiency of data processing. Attached Figure Description

[0041] Figure 1 This is a flowchart of a data processing method based on a search engine provided in an embodiment of this application;

[0042] Figure 2 yes Figure 1 The flowchart of step S102 in the document;

[0043] Figure 3 yes Figure 1 The flowchart of step S103 in the process;

[0044] Figure 4 yes Figure 1 The flowchart of step S104 in the process;

[0045] Figure 5 yes Figure 4 The flowchart of step S403 in the process;

[0046] Figure 6 yes Figure 4 The flowchart of step S402 in the document;

[0047] Figure 7 This is a flowchart of a data processing method based on a search engine provided in another embodiment of this application;

[0048] Figure 8 This is a schematic diagram of the structure of the data processing device based on a search engine provided in the embodiments of this application;

[0049] Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0051] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0053] First, let's analyze some of the terms used in this application:

[0054] Natural Language Processing (NLP): NLP uses computers to process, understand, and utilize human language (such as Chinese and English). NLP is a branch of artificial intelligence and an interdisciplinary field of computer science and linguistics, often referred to as computational linguistics. NLP includes syntactic analysis, semantic analysis, and discourse understanding. It is commonly used in machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, intent recognition, information extraction and filtering, text classification and clustering, sentiment analysis, and opinion mining. It involves data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computation.

[0055] Euclidean distance: In mathematics, Euclidean distance, or Euclidean metric, is the "ordinary" (i.e., straight-line) distance between two points in Euclidean space. Using this distance, Euclidean space is called a metric space. The associated norm is called the Euclidean norm. Earlier literature referred to it as the Pythagorean metric. Euclidean distance is a commonly used definition of distance, referring to the true distance between two points in m-dimensional space, or the natural length of a vector (i.e., the distance from that point to the origin). In two-dimensional and three-dimensional space, the Euclidean distance is simply the actual distance between two points.

[0056] Cosine Function: The cosine function generally refers to the cosine, a type of trigonometric function. In right triangle ABC, ∠C = 90°, the cosine of ∠A is the ratio of its adjacent side to the hypotenuse of the triangle, i.e., cosA = b / c, which can also be written as cosA = AC / AB. The cosine function is: f(x) = cosx (x∈R).

[0057] Unweighted Pair-Group Method with Arithmetic Means (UPGMA) is a commonly used clustering analysis method, originally developed for solving classification problems. When used to reconstruct phylogenetic trees, it assumes that divergent rates occur the same number of times in each lineage during evolution, meaning the rate of nucleotide or amino acid substitution is equal and constant. The phylogenetic tree generated by UPGMA can be considered a simplified representation of a species tree; after each divergence, the branches from the common ancestor node to the two operational taxonomic units (OTUs) are of equal length. Therefore, this method is frequently used for reconstructing species trees.

[0058] Information extraction is a text processing technique that extracts factual information such as entities, relationships, and events from natural language text and outputs it as structured data. Information extraction is a technique for extracting specific information from text data. Text data is composed of specific units, such as sentences, paragraphs, and chapters. Text information is composed of smaller, specific units, such as characters, words, phrases, sentences, paragraphs, or combinations of these units. Extracting noun phrases, names of people, and place names from text data is an example of text information extraction. Of course, text information extraction techniques can extract information of various types.

[0059] Edit distance: Edit distance is mainly used to compare the similarity between two strings. In NLP tasks, it is common to compare the similarity between two strings, such as spell correction and reference judgment.

[0060] User personas, also known as user roles, are an effective tool for outlining target users and connecting user needs with design direction. They are widely used across various fields. In practice, we often use the simplest and most relatable language to connect user attributes, behaviors, and expectations with data. As virtual representatives of actual users, user personas are not constructed outside of the product and market; they need to be representative, representing the product's primary audience and target group.

[0061] Baseline drift: Baseline drift is a metrology term introduced in 2015, referring to the deviation of the detector baseline in a single direction over time.

[0062] Based on this, embodiments of this application provide a data processing method and apparatus, electronic device and storage medium based on a search engine, which can selectively analyze the semantics of user query statements within a preset time, thereby improving the efficiency of data analysis.

[0063] The data processing method, apparatus, electronic device, and storage medium based on a search engine provided in this application are specifically described through the following embodiments. First, the data processing method based on a search engine in this application is described.

[0064] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0065] The fundamental technologies of artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly include computer vision, robotics, biometrics, search engine-based data processing, natural language processing, and machine learning / deep learning.

[0066] The data processing method based on a search engine provided in this application relates to the field of artificial intelligence technology. This data processing method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the data processing method based on a search engine, but is not limited to the above forms.

[0067] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0068] Figure 1 This is an optional flowchart of a data processing method based on a search engine provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S101 to S105.

[0069] Step S101: Obtain user information from multiple users based on a search engine;

[0070] It should be noted that user information includes the user's query information and the time when the query occurred.

[0071] In step S101 of some embodiments, user information of multiple different users is obtained based on a search engine. The user information includes user identifiers, query information, the time when the query information occurred, semantic vectors, and other information, which facilitates subsequent calculation of user information.

[0072] It should be noted that search engines can be full-text search engines, directory search engines, meta search engines, or vertical search engines, etc. Specifically, search engines can be Baidu search engine, Google search engine, Google search engine, etc.

[0073] Understandably, the query information can be a question entered by the user, such as "How to choose insurance?", "How to select insurance?", "What types of insurance are there?", etc. Each character in the query information has a specific semantic vector, and there is a one-to-one correspondence between the character information and the semantic vector.

[0074] Step S102: Divide the query information according to the preset time rules and the time of occurrence to obtain multiple query sets;

[0075] In step S102 of some embodiments, the query information is divided according to a preset time rule and the occurrence time of the query information to obtain multiple query sets, which facilitates the subsequent calculation of the correction tendency value and improves the calculation efficiency.

[0076] It should be noted that the preset time rule can be to divide the inquiry information according to a fixed time period, according to any time period, or according to a periodic time period. For example, when the preset time rule is to divide the inquiry information according to a fixed time period, then all inquiry information within that time period is divided from 8:00 AM to 10:00 AM every day; when the preset time rule is to divide the inquiry information according to any time period, then all inquiry information within that time period can be divided from 8:00 AM to 1:00 PM on Thursday and from 12:00 PM to 3:00 PM on Friday; when the preset time rule is to divide the inquiry information according to a periodic time period, then all inquiry information within that time period is divided from 10:00 AM to 2:00 PM every Tuesday, etc. This embodiment does not impose specific limitations.

[0077] Step S103: Perform semantic space calculation on the query set to obtain the correction tendency value of the query set;

[0078] In step S103 of some embodiments, semantic space calculation is performed on the query set obtained in step S102 to obtain the correction tendency value of the query set, which facilitates the determination of the degree of deviation of the semantic vector in the query set. Obtaining the correction tendency value of each query set can accurately determine whether the semantics of different query sets are similar, which is convenient for subsequent adjustments to the search engine.

[0079] It should be noted that for each set of queries, if the query statements in the set are very similar, it indicates that the user's tendency to correct the query is more obvious, that is, the larger the correction tendency value is.

[0080] Step S104: Filter the correction tendency value according to the preset drift value to obtain the target correction tendency value;

[0081] In step S104 of some embodiments, the correction tendency value is filtered according to the preset drift value to obtain the target correction tendency value, thereby facilitating the subsequent adjustment of the search engine.

[0082] It should be noted that, since each user has different language habits and typing methods, the preset drift value is different for each user and changes with the user's language habits. The setting of the preset drift value is used to solve the baseline drift caused by user differences, and this embodiment does not impose specific limitations.

[0083] Step S105: Adjust the search engine according to the target correction tendency value.

[0084] In step S105 of some embodiments, the search engine is adjusted according to the target correction tendency value, thereby optimizing the search engine algorithm and improving the user's search efficiency.

[0085] Steps S101 to S105 as shown in the embodiments of this application first obtain user information of multiple users based on the search engine, and divide the query information according to preset time rules and the occurrence time of the query information in the user information, thereby obtaining multiple query sets divided by time, realizing the division of query information within a preset time. Then, semantic space calculation is performed on the query sets to obtain the correction tendency value of each query set, and the correction tendency value is filtered according to a preset drift value, thereby selecting the target correction tendency value from multiple correction tendency values. Finally, the search engine is adjusted according to the target correction tendency value, thereby enabling selective optimization of the search engine and improving the efficiency of data processing.

[0086] Please see Figure 2 In some embodiments, step S102 may include, but is not limited to, steps S201 to S203:

[0087] Step S201: Sort the query information according to the occurrence time to obtain the occurrence time sequence;

[0088] In step S201 of some embodiments, the query information is sorted according to the occurrence time to obtain the event sequence, which facilitates the subsequent division of time intervals.

[0089] It should be noted that the query information can be sorted according to the time of occurrence, either from early to late or from late to early. For example, if the query information occurs at 8:00 AM, 8:10 AM, 8:20 AM, 12:00 PM, 1:00 PM, 3:00 PM, or 2:00 PM, the query information can be arranged in chronological order to obtain a time series of 8:00 AM, 8:10 AM, 8:20 AM, 12:00 PM, 1:00 PM, 2:00 PM, and 3:00 PM, or a time series of 3:00 PM, 2:00 PM, 1:00 PM, 12:00 PM, 8:20 PM, 8:10 PM, and 8:00 PM. Each time point in the time series corresponds to a query information. This embodiment does not impose specific limitations.

[0090] Step S202: Divide the occurrence time series according to preset time rules to obtain multiple time intervals;

[0091] In step S202 of some embodiments, the occurrence time sequence is divided according to a preset time rule to obtain multiple time intervals, which facilitates the subsequent statistical analysis of query information in all time intervals.

[0092] It should be noted that dividing the occurrence time sequence according to the preset time rule means clustering the query information into time periods according to the density of the query information. In this embodiment, the preset time rule is the time interval between the occurrence time of a query information and the occurrence time of the previous query information. For example, if the time interval between the occurrence time of a query information and the occurrence time of the previous query information is less than one minute, it is classified into the previous time interval; otherwise, it is classified into the next time interval, and so on, to obtain multiple time intervals. The time interval can be set by the user according to their needs, such as two minutes, five minutes, or six minutes, etc. This embodiment does not impose specific restrictions.

[0093] Step S203: Statistically analyze all query information within the time interval to obtain multiple query sets.

[0094] In step S203 of some embodiments, all query information within a time interval is statistically analyzed to obtain a query set for each time interval, which facilitates subsequent spatial semantic calculations and thus accurately determines the user's search intention within a certain time interval.

[0095] Please see Figure 3 In some embodiments, step S103 may include, but is not limited to, steps S301 to S304:

[0096] It should be noted that the query set includes multiple semantic vectors.

[0097] Step S301: Obtain multiple semantic vectors from the query set;

[0098] It should be noted that the query set includes queries from multiple users, and the semantic vectors of different characters in each query are different. For example, if the query is "How to choose insurance?", the semantic vector corresponding to each character is "3.23,543.1,43.3,5,3.1", which can improve the accuracy of calculating the propensity score.

[0099] Step S302: Calculate the semantic vector based on the preset distance function to obtain the distance parameters of the semantic vector;

[0100] It should be noted that the preset distance function can be the Euclidean distance function, the cosine function, the edit distance function of the vector corresponding to the question, etc., and this embodiment does not impose specific restrictions.

[0101] Step S303: The distance parameters are averaged according to the clustering analysis algorithm to obtain the semantic space value;

[0102] It should be noted that the clustering analysis algorithm is the unweighted group average method.

[0103] Step S304: Obtain the correction tendency value of the query set based on the semantic space value and the number of semantic vectors.

[0104] Steps S301 to S304 as shown in the embodiments of this application first obtain multiple semantic vectors in the query set, then calculate the semantic vectors based on a preset distance function to obtain the distance parameters of the semantic vectors, thereby determining the semantic distance between adjacent semantics. Then, the distance parameters are averaged according to a clustering analysis algorithm to obtain the semantic space value, that is, the size of the semantic space. Finally, the correction tendency value of the query set is obtained based on the semantic space value and the number of semantic vectors, thereby completing the calculation for each query set, which is convenient for determining whether the keywords searched by the user are adjacent or similar.

[0105] It should be noted that each unit time interval contains multiple query sets, and each query set contains several users' query statements. If the semantics of these users' queries are very similar, it indicates that the user's tendency to "correct the query" is more obvious. That is, if the semantics are close enough, we have caught a bad case. When a user cannot get the desired search results even after entering several semantically similar query statements, it is necessary to calculate the size of the semantic space formed by the query sets in each time interval. The smaller the size, the more likely it is to be a bad case. That is, the correction tendency value is the reciprocal of the semantic space value of the user in a certain time interval. The specific correction tendency value algorithm is shown in the following formula (1):

[0106]

[0107] It is understandable that set C is the semantic vector corresponding to all queries of a certain user within a certain unit time interval, ‖C‖ represents the number of elements in set C, function d is the distance function between two vectors, and (a,b) are any two distance parameters in function d.

[0108] It should be noted that the distance function is the same as the preset distance function mentioned above, and will not be repeated here.

[0109] Please see Figure 4 In some embodiments, step S104 may include, but is not limited to, steps S401 to S403:

[0110] Step S401: Generate a tendency curve based on the corrective tendency value and the time of occurrence;

[0111] In step S401 of some embodiments, a trend curve of occurrence time-correction trend value is generated for all query sets with the correction trend value as the vertical axis and the occurrence time as the horizontal axis, which facilitates the filtering of points with excessively large trend values.

[0112] Step S402: Divide the tendency curve according to the preset drift value to obtain the correction threshold corresponding to the correction tendency value;

[0113] In step S402 of some embodiments, the tendency curve is divided according to a preset drift value to obtain a correction threshold corresponding to the correction tendency value, thereby realizing correction according to the search habits of different users and improving the user experience.

[0114] It should be noted that, since each user has different input semantic habits and search habits, different users have different preset drift values. For example, some users are used to inputting semantically similar sentences but not correcting them, while others are used to inputting semantically dissimilar questions but correcting them. Therefore, different preset drift values ​​need to be set for each user.

[0115] Step S403: Filter the correction tendency values ​​based on the correction threshold to obtain the target correction tendency value.

[0116] In step S403 of some embodiments, the correction tendency value is screened based on the correction threshold to determine the target correction tendency value, which facilitates subsequent optimization of the search engine based on the target correction tendency value and solves the baseline drift caused by user differences.

[0117] Please see Figure 5 In some embodiments, step S403 may also include, but is not limited to, steps S501 to S502:

[0118] Step S501: Compare the correction threshold with the correction tendency value;

[0119] Step S502: When the comparison result shows that the correction tendency value is greater than the correction threshold, the correction threshold is determined as the target tendency value.

[0120] In the embodiments of this application, steps S501 to S502 involve comparing the correction threshold with the correction tendency value. If the comparison result shows that the correction tendency value is greater than the correction threshold, the correction threshold is determined as the target tendency value. If the comparison result shows that the correction tendency value is less than or equal to the correction threshold, then it is not necessary to filter the correction tendency values ​​that are less than the correction threshold, and the correction tendency value remains unchanged, thereby solving the baseline drift caused by user differences.

[0121] Please see Figure 6 In some embodiments, step S402 includes, but is not limited to, steps S601 to S602:

[0122] It should be noted that the preset drift value includes the drift variance and the drift mean.

[0123] Step S601: Calculate the tendency curve based on the drift average value;

[0124] Step S602: Discretize the calculated tendency curve based on the drift variance to obtain the correction threshold corresponding to the correction tendency value.

[0125] In the embodiments of this application, steps S601 to S602 involve calculating the tendency curve based on the drift average, subtracting the drift average of each user from the tendency curve to eliminate the differences between different users, then superimposing all the tendency curves for analysis, and finally discretizing the calculated tendency curve based on the drift variance to obtain the correction threshold corresponding to the correction tendency value. The drift variance can reflect the degree of deviation of the tendency curve, thereby enabling correction for different users.

[0126] It is understood that in this embodiment, the method of drift variance plus drift average is used to calculate the correction threshold for each user. Since each user has different habits, the threshold for each user is different. The method of drift variance plus drift average can calculate the threshold for each user more accurately.

[0127] Figure 7 This is a flowchart of a data processing method based on a search engine provided in another embodiment of this application. Figure 7 The method may include, but is not limited to, steps S701 to S702.

[0128] Step S701: Monitor the correction tendency value based on preset monitoring rules to determine the mutation tendency value;

[0129] In step S701 of some embodiments, the correction tendency value is detected based on preset monitoring rules, thereby realizing real-time monitoring of the calculation process of the correction tendency value and obtaining the mutation tendency value, which facilitates subsequent updates and optimizations of the search engine.

[0130] Step S702: Update the search engine based on the mutation tendency value.

[0131] In step S702 of some embodiments, the mutation tendency value is determined in step S701, and the search engine algorithm corresponding to the mutation tendency value is determined based on the mutation tendency value. Finally, the search engine is updated based on the mutation tendency value, thereby optimizing the algorithm in the search engine.

[0132] It should be noted that, since there are a large number of users of search engines, it is necessary to design preset monitoring rules to calculate and monitor the correction tendency value of all users in real time. When the correction tendency value is abnormal, a warning will be issued, thereby optimizing the search engine algorithm.

[0133] It's understandable that there could be many reasons for an abnormality in the corrected propensity score. Although each user has different search habits, their input behavior should be stable. Therefore, a sudden increase in the propensity score must be caused by an external force. For example, if the search algorithm in a search engine is modified, causing some users' corrected propensity scores to rise sharply, it means that the user is dissatisfied with the effect of the new search algorithm. This is when the abnormality in the corrected propensity score occurs. Among these reasons, a user's corrected propensity score may rise sharply because different keywords produce equally few results, or the search results are unrelated to the keywords.

[0134] It's worth noting that some users habitually input semantically similar queries without actually performing input correction. However, if several consecutive queries they input are sufficiently similar (e.g., they only changed one character), an alert should still be triggered. This prevents key bad cases from being missed, thereby improving analysis efficiency.

[0135] In some embodiments, an alert is issued when a user exhibits a sudden change in their profile, facilitating timely detection of changes in the user profile and allowing for adjustments to the search engine's algorithm strategy.

[0136] It should be noted that search algorithms generally retain characteristic data for each user, such as the user's interests, intended insurance information, etc. This characteristic data is integrated into a user profile, and then different users are given search results that are tailored to their habits, thereby achieving a better user experience. A sudden increase in a user's correction tendency value may be due to the user adding some new interests, in which case the user profile needs to be updated so that the search engine can keep the synchronization with the user's characteristics.

[0137] Please see Figure 8 This application also provides a data processing apparatus based on a search engine, which can implement the above-described data processing method based on a search engine. The apparatus includes:

[0138] The information acquisition module 801 is used to acquire user information of multiple users based on a search engine. The user information includes the user's query information and the time when the query information occurred.

[0139] The semantic segmentation module 802 is used to segment query information according to preset time rules and occurrence time to obtain multiple query sets;

[0140] The spatial computing module 803 is used to perform semantic space computing on the query set to obtain the correction tendency value of the query set;

[0141] The target screening module 804 is used to screen the correction tendency value according to the preset drift value to obtain the target correction tendency value;

[0142] Engine tuning module 805 is used to adjust the search engine based on the target correction tendency value.

[0143] The specific implementation of this search engine-based data processing device is basically the same as the specific implementation of the search engine-based data processing method described above, and will not be repeated here.

[0144] This application also provides an electronic device, which includes: a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for communication between the processor and the memory. When the program is executed by the processor, it implements the aforementioned search engine-based data processing method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0145] Please see Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0146] The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0147] The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 using the search engine-based data processing method of the embodiments of this application.

[0148] The input / output interface 903 is used to implement information input and output;

[0149] The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0150] Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904);

[0151] The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0152] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described search engine-based data processing method.

[0153] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0154] The data processing method, apparatus, electronic device, and storage medium based on a search engine provided in this application first obtain user information from multiple users through a search engine, and then divide the query information according to preset time rules and the occurrence time of the query information in the user information, thereby obtaining multiple query sets divided by time, realizing the division of query information within a preset time. Then, semantic space calculation is performed on the query sets to obtain the correction tendency value of each query set, and the correction tendency value is filtered according to a preset drift value, thereby selecting the target correction tendency value from multiple correction tendency values. Finally, the search engine is adjusted according to the target correction tendency value, thereby enabling selective optimization of the search engine and improving the efficiency of data processing.

[0155] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0156] It will be understood by those skilled in the art that Figure 1-7 The technical solutions shown do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0157] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0158] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0159] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0160] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0161] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0162] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0163] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0164] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0165] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A data processing method based on a search engine, characterized by, The method comprises: obtaining user information of a plurality of users based on a search engine, wherein the user information comprises inquiry information of the users and occurrence time of the inquiry information; dividing the inquiry information according to a preset time rule and the occurrence time to obtain a plurality of inquiry sets; performing semantic space calculation on the inquiry sets to obtain a correction tendency value of the inquiry sets; screening the correction tendency value according to a preset drift value to obtain a target correction tendency value; adjusting the search engine according to the target correction tendency value; wherein the inquiry set comprises a plurality of semantic vectors; the semantic space calculation on the inquiry set to obtain the correction tendency value of the inquiry set comprises: obtaining a plurality of semantic vectors in the inquiry set; calculating the semantic vectors based on a preset distance function to obtain distance parameters of the semantic vectors; performing average calculation on the distance parameters according to a clustering analysis algorithm to obtain a semantic space value; obtaining the correction tendency value of the inquiry set according to the semantic space value and the number of the semantic vectors; the screening of the correction tendency value according to the preset drift value to obtain the target correction tendency value comprises: generating a tendency curve according to the correction tendency value and the occurrence time; dividing the tendency curve according to the preset drift value to obtain a correction threshold corresponding to the correction tendency value; screening the correction tendency value based on the correction threshold to obtain the target correction tendency value.

2. The search engine based data processing method of claim 1, wherein, the dividing of the inquiry information according to the preset time rule and the occurrence time to obtain a plurality of inquiry sets comprises: sorting the inquiry information according to the occurrence time to obtain an occurrence time sequence; dividing the occurrence time sequence according to the preset time rule to obtain a plurality of time intervals; counting all the inquiry information in the time interval to obtain a plurality of inquiry sets.

3. The search engine based data processing method of claim 1, wherein, the screening of the correction tendency value based on the correction threshold to obtain the target correction tendency value comprises: comparing the correction threshold with the correction tendency value; when the comparison result is that the correction tendency value is greater than the correction threshold, determining that the correction threshold is a target tendency value.

4. The search engine based data processing method of claim 1, wherein, the preset drift value comprises a drift variance and a drift average value; the dividing of the tendency curve according to the preset drift value to obtain a correction threshold corresponding to the correction tendency value comprises: calculating the tendency curve according to the drift average value; performing discrete processing on the calculated tendency curve according to the drift variance to obtain the correction threshold corresponding to the correction tendency value.

5. The search engine based data processing method of claim 1, wherein, further comprising: monitoring the correction tendency value based on a preset monitoring rule to determine a mutation tendency value; updating the search engine according to the mutation tendency value.

6. A data processing apparatus based on a search engine, characterized in that, the device comprises: an information acquisition module configured to obtain user information of a plurality of users based on a search engine, wherein the user information comprises inquiry information of the users and occurrence time of the inquiry information; The semantic division module is configured to divide the query information according to a preset time rule and the occurrence time, and obtain a plurality of query sets. The spatial calculation module is configured to perform semantic spatial calculation on the query sets, and obtain a correction tendency value of the query sets. The target screening module is configured to screen the correction tendency value according to a preset drift value, and obtain a target correction tendency value. The engine adjustment module is configured to adjust the search engine according to the target correction tendency value. The query sets include a plurality of semantic vectors. The semantic spatial calculation on the query sets includes: obtaining a plurality of semantic vectors in the query sets; calculating the semantic vectors based on a preset distance function, and obtaining distance parameters of the semantic vectors; performing average calculation on the distance parameters according to a clustering analysis algorithm, and obtaining semantic spatial values; obtaining the correction tendency value of the query sets according to the semantic spatial values and the number of the semantic vectors; The screening of the correction tendency value according to the preset drift value includes: generating a tendency curve according to the correction tendency value and the occurrence time; dividing the tendency curve according to the preset drift value, and obtaining a correction threshold corresponding to the correction tendency value; screening the correction tendency value based on the correction threshold, and obtaining the target correction tendency value.

7. An electronic device, comprising: The electronic device includes a memory and a processor. The memory stores a computer program. The processor executes the computer program to implement the search engine-based data processing method of any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, the computer-readable storage medium comprising: The computer program is executed by the processor to implement the search engine-based data processing method of any one of claims 1 to 5.

Images