Alarm information processing method and device and storage medium

Abstract

The invention discloses an alarm information processing method and device and a storage medium, and relates to the technical field of data processing.The method comprises the steps that an unstructured alarm text is converted into an accurate keyword set through word segmentation processing, the limitation of traditional simple text matching is broken through, alarm core semantic information is effectively extracted, and redundant interference is reduced; according to the method, the mapping relation between the keyword and the alarm identifier is established and stored in the database, so that an efficient retrieval structure based on the inverted index is realized, the overhead of full-text scanning is avoided, and the retrieval speed of massive alarm data is remarkably improved; according to the method, the target alarm text can be more accurately positioned based on semantic relevance of the keywords, and the rigidity of traditional fixed condition retrieval is overcome, so that the practicability of alarm information retrieval is synchronously optimized in the aspects of efficiency and precision. The technical problem of low retrieval efficiency in a traditional retrieval mode is solved, and the technical effect of improving the retrieval speed of mass alarm data is achieved.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to a method, device, and storage medium for processing alarm information. Background Art

[0002] In modern information technology systems, alarm information, as an important part of the monitoring system, undertakes the key task of promptly detecting and responding to potential faults or anomalies. The current common traditional retrieval method mainly stores the alarm information collected by the monitoring system or log tool in a relational database and realizes retrieval analysis through fixed retrieval conditions set by users. However, this traditional retrieval method relies on simple text matching and has two problems. One is that it cannot accurately evaluate the similarity of alarm content, resulting in limited accuracy and practicality of retrieval results. The other is that as the system scale expands and the alarm information surges, the problem of low efficiency gradually emerges. Summary of the Invention

[0003] This application provides a method, device, and storage medium for processing alarm information, which at least solves the technical problems of low retrieval efficiency and insufficient accuracy in the related art, and achieves the technical effect of improving the retrieval speed of massive alarm data.

[0004] This application provides a method for processing alarm information, including: receiving alarm text data; performing word segmentation processing on the alarm text data to obtain at least one alarm keyword; establishing a mapping relationship between the alarm keyword and the identification information of the alarm text data; storing the mapping relationship in a preset database, so as to retrieve the corresponding target alarm text data from the preset database according to the target alarm keyword and the mapping relationship.

[0005] This application provides a method for processing alarm information, including: obtaining a target alarm keyword input by a user; calling the mapping relationship stored in a preset database, where the mapping relationship stores the mapping relationship between the alarm keyword and the identification information of the alarm text data; determining target identification information according to the mapping relationship, and determining the target alarm text data according to the target identification information.

[0006] This application also provides an electronic device, including: a memory for storing a computer program; a processor for implementing the steps of any of the above methods for processing alarm information when executing the computer program.

[0007] This application also provides a computer-readable storage medium, in which a computer program is stored, and when the computer program is executed by a processor, the steps of any of the above methods for processing alarm information are implemented.

[0008] The present application also provides a computer program product, including a computer program, which implements the steps of any of the above-mentioned warning information processing methods when executed by a processor.

[0009] Through the present application, the unstructured warning text is converted into a set of accurate keywords through word segmentation processing, breaking through the limitations of traditional simple text matching, effectively extracting the core semantic information of the warning, and reducing redundant interference; by establishing the mapping relationship between keywords and warning identifiers and storing them in the database, an efficient retrieval structure based on inverted index is realized, avoiding the overhead of full-text scanning, and significantly improving the retrieval speed of massive warning data; based on the semantic relevance of keywords, the target warning text can be located more accurately, overcoming the rigidity of traditional fixed-condition retrieval, thus synchronously optimizing the practicality of warning information retrieval in terms of efficiency and accuracy. It solves the technical problems of low retrieval efficiency and insufficient accuracy existing in the traditional retrieval method, and achieves the technical effect of improving the retrieval speed of massive warning data. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0011] Figure 1 It is a flowchart of a method for processing warning information provided by an embodiment of the present application;

[0012] Figure 2 It is a schematic diagram of word segmentation processing provided by an embodiment of the present application;

[0013] Figure 3 It is a schematic diagram of a mapping relationship provided by an embodiment of the present application;

[0014] Figure 4 It is a specific flowchart of a method for processing warning information provided by an embodiment of the present application;

[0015] Figure 5 It is a flowchart of another method for processing warning information provided by an embodiment of the present application;

[0016] Figure 6 It is a simplified flowchart of the storage and retrieval of warning information provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the protection scope of the present application.

[0018] It should be noted that in the description of the present application, the terms "include", "comprise" or any other variation thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. The terms "first", "second", etc. in the present application are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0019] In order to enable those skilled in the art of the present technology to better understand the solution of the present application, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0020] As Figure 1 shown, the present application provides a method for processing alarm information, including:

[0021] S11: Receive alarm text data.

[0022] Specifically, as the initial link of the alarm information processing flow, it is necessary to construct a data input interface to obtain the original alarm information and provide basic data support for subsequent processing. This step can achieve the capture and access of alarm text data in different forms and from different sources by setting a standardized data reception mechanism, ensuring that various alarm text data can be effectively identified and processed. The core lies in completing the transfer process of alarm text data from the external environment to the processing system, focusing on the integrity of the data and the compatibility of the format. By adapting different data input protocols or interfaces, the scattered alarm text data is aggregated into the processing system to form a unified data input source for subsequent word segmentation processing, laying a data foundation for the entire alarm information processing flow and ensuring that subsequent operations such as keyword extraction and mapping relationship establishment can be carried out based on accurate and complete original data.

[0023] S12: Perform word segmentation processing on the alarm text data to obtain at least one alarm keyword.

[0024] Specifically, as the transformation link in the alarm information processing flow, it mainly lies in using the word segmentation technology in natural language processing to transform unstructured alarm text data into structured key information units, providing a semantic basis for subsequent data processing. Specifically, in this step, through a text parsing mechanism, the input alarm text data is segmented into language units, and the words or phrases carrying the core semantics (i.e., alarm keywords) are identified and extracted. Redundant characters or invalid words without substantial meaning can be removed, so that the unstructured natural language text is transformed into discrete semantic units that can be efficiently processed by computer systems. Its essence is to extract the key features that can accurately describe the alarm content from the original text through the division of semantic granularity, forming the core data elements used to represent alarm information, thereby laying a foundation for establishing the mapping relationship between keywords and alarm records and realizing efficient semantic-based retrieval in the subsequent process, ensuring that the system can accurately locate and match relevant alarm information through keywords, and improving the pertinence and effectiveness of the overall processing flow.

[0025] S13: Establish a mapping relationship between the alarm keywords and the identification information of the alarm text data.

[0026] Specifically, as the link to realize data association in the alarm information processing flow, it mainly lies in establishing a structured data mapping mechanism to semantically associate and bind the alarm keywords obtained by word segmentation with the corresponding alarm text data, providing an indexing basis for subsequent efficient retrieval.

[0027] Specifically, by defining a unique identification information (such as the ID of the alarm text data) as the data anchor point, a two-way correspondence is established between the alarm keywords with semantic representation and the text data carrying specific alarm content, making each keyword an index entry for retrieving the target alarm text, while the identification information serves as a pointer to the specific data storage location. Its essence is to transform unstructured text information into structured index data. Through the construction of this mapping relationship, it is possible to bypass the sequential scanning of all alarm text data and directly locate the associated alarm text data quickly through keyword matching, thereby building an efficient retrieval path at the data layer and laying a technical foundation for realizing fast query and accurate matching based on keywords.

[0028] S14: Store the mapping relationship in a preset database to retrieve the corresponding target alarm text data from the preset database according to the target alarm keywords and the mapping relationship.

[0029] Specifically, by storing the mapping relationship between the previously established alarm keywords and alarm text identification information in a structured manner, a stable data support platform is provided for efficient keyword-based retrieval in the future. This step stores the mapping relationship in a preset database in a specific data format (such as key-value pairs, index tables, etc.), enabling the formation of an associated data set that can be quickly read and queried by a computer between discrete keywords and corresponding alarm records. Its essence is to build an index bridge from "semantic alarm keywords" to "specific alarm text data" at the data layer. When retrieving later, instead of traversing all alarm text data, directly through the query mechanism of the preset database, according to the target keyword to match the mapping relationship, quickly locate and extract the corresponding identification information, and then associate with the specific alarm text data, so as to achieve efficient data interaction from the storage layer to the application layer, providing stable and reliable underlying data support for real-time retrieval and analysis.

[0030] In summary, through word segmentation processing, the unstructured alarm text is converted into a precise set of keywords, breaking through the limitations of traditional simple text matching, effectively extracting the core semantic information of the alarm, and reducing redundant interference; by establishing the mapping relationship between keywords and alarm identifiers and storing them in the database, an efficient retrieval structure based on inverted index is realized, avoiding the overhead of full-text scanning, and significantly improving the retrieval speed of massive alarm data; based on the semantic relevance of keywords, the target alarm text can be located more accurately, overcoming the rigidity of traditional fixed-condition retrieval, thus synchronously optimizing the practicality of alarm information retrieval in terms of efficiency and accuracy. It solves the technical problems of low retrieval efficiency and insufficient accuracy existing in traditional retrieval methods, and achieves the technical effect of improving the retrieval speed of massive alarm data.

[0031] Such as Figure 4 , in an exemplary embodiment, word segmentation processing is performed on the alarm text data to obtain at least one alarm keyword, including: loading a variety of dictionary resources, matching the alarm text data with the variety of dictionary resources; performing word segmentation processing on the alarm text data according to the keywords in the variety of dictionary resources to obtain at least one alarm keyword.

[0032] Specifically, by constructing a multi-level dictionary resource system, comprehensive vocabulary matching support is provided for the semantic analysis of alarm text data, so as to achieve precise segmentation and key information extraction of complex text content.

[0033] Specifically, by loading multiple dictionary resources covering different semantic categories (such as collections of basic language, domain-specific, user-defined vocabulary, etc.), the alarm text data is structurally matched with the standard vocabulary in each dictionary. Based on the preset lexical units in the dictionary, the text is divided into semantic granularity, enabling unstructured natural language text to be split into alarm keywords with actual semantics according to predefined effective lexical boundaries. Its essence is to utilize the collaborative matching mechanism of multi-source dictionaries to break through the lexical coverage limitations of a single dictionary. Through multi-level and multi-dimensional lexical comparison, it ensures that different types of language elements (common vocabulary, technical terms, user-defined vocabulary, etc.) in the text can be accurately identified and extracted, forming a keyword set containing complete semantic information.

[0034] The reason for selecting multiple dictionary resources is to construct a more adaptable word segmentation system. For example, it can include the following three dictionaries. One is a general dictionary, which ensures the standardized parsing of basic language elements by collecting common vocabulary in the general language, laying the foundation for the universality of word segmentation. The second is a user-defined dictionary, which allows users to dynamically expand the vocabulary library according to specific scenario requirements (such as emerging industry terms, Internet buzzwords, etc.), and can flexibly adapt to the changing language environment and personalized application needs. The third is a professional dictionary, which focuses on the exclusive terms in the field of monitoring and alarm, ensuring the accurate understanding of industry-specific semantics and avoiding word segmentation deviations caused by differences in domain vocabulary. This multi-dictionary collaborative query mechanism, by covering the full-dimensional lexical space of "general + professional + customized", not only significantly improves the accuracy and integrity of the word segmentation results, but also endows powerful environmental adaptability and functional extensibility through the modular dictionary expansion design, enabling it to effectively serve diverse user groups and complex application scenarios, fundamentally solving the deficiencies of a single dictionary in lexical coverage and scenario adaptability.

[0035] Such as Figure 2As shown, in an exemplary embodiment, the alarm text data is segmented according to keywords in multiple dictionary resources to obtain at least one alarm keyword, including: starting from the current starting position of the alarm text data, scanning character by character towards the end position, and initializing the scanning window to a preset maximum length; matching the string within the scanning window with multiple dictionary resources; if the match is successful, extracting the longest string matched by the current scanning window as the alarm keyword for segmentation processing, and moving the scanning pointer to the next character position after the end of the longest string; if the match fails, gradually reducing the scanning window, and determining whether the number of characters in the scanning window is greater than 1. If it is greater than 1, return to the step of matching the string within the scanning window with multiple dictionary resources to perform the matching again until the match is successful or until the number of characters in the scanning window is equal to 1. Taking the string or single character that matches successfully within the scanning window as the alarm keyword for segmentation processing, and moving the scanning pointer to the next character position; updating the next character position to the new current starting position, and returning to the step of starting from the current starting position of the alarm text data, scanning character by character towards the end position, and initializing the scanning window to a preset maximum length according to the new starting position to continue matching the remaining alarm text data except for single characters until the segmentation processing of all characters in the alarm text data is completed.

[0036] Specifically, in this embodiment, through a dynamic window matching method, combined with multiple dictionary resources, the alarm text data is split into meaningful alarm keywords. Its essence is to use a variable-length scanning window to start from the text starting position and find the longest string that matches it in the dictionary resources to determine the segmentation result.

[0037] First, starting from the starting position of the alarm text, using the preset maximum length as the scanning window, scan the text character by character from the starting position to the end position. The string within the scanning window will be matched with multiple dictionary resources. The purpose of doing this is to give priority to trying to find a longer matching string because in natural language, longer words can often express semantics more accurately. If the match is successful, extract this longest matching string as an alarm keyword, and then move the scanning pointer to the next position after the end of the string to continue the subsequent matching.

[0038] If the match fails, gradually reduce the length of the scanning window and try to match again until a matching string is found or the scanning window is reduced to only one character. When the scanning window is reduced to one character, if the character can be successfully matched in the dictionary, take it as a keyword. Then, update the current starting position and repeat the above scanning and matching process until the entire alarm text data is processed.

[0039] It should be understood that the character-by-character scanning of the text from the starting position to the ending position can be a forward scan from left to right, a reverse scan from right to left, or a bidirectional scan that combines the advantages of both and synchronously performs the longest match in both the forward and reverse directions. Specifically, the forward scan from left to right starts from the starting position of the text and expands the match character by character to the right through a preset maximum length, preferentially identifying long word items to reduce ambiguity (for example, for "Server memory severe alarm", "Server memory" is matched first instead of being split into "Server" and "Memory"), which conforms to the reading habits and vocabulary composition logic of most languages from left to right; the reverse scan from right to left starts from the ending position of the text and backtracks to the left for matching; while the bidirectional scan combines the advantages of both and synchronously performs the longest match in both the forward and reverse directions, further eliminating word segmentation ambiguity through result comparison (for example, bidirectional scanning can avoid incorrect segmentation caused by the order of the word library or the text structure in unidirectional matching).

[0040] Regardless of the scanning direction adopted, its essence is to disassemble the continuous text stream into discrete keywords that conform to semantic rules through a structured character traversal strategy and combine the vocabulary matching capabilities of multiple dictionary resources. This multi-directional scanning mechanism effectively improves the system's adaptability to complex text structures (such as a mixture of Chinese and English, combinations of domain terms) by covering different character processing orders, ensuring that the core semantic units in the alarm text can be accurately extracted in different language environments and application scenarios, providing high-quality basic data for subsequent index construction and efficient retrieval.

[0041] Suppose the alarm text data to be segmented is "Server memory severe alarm", and the dictionary is: {"Server", "Memory", "Server memory", "Alarm", "Severe"}. The word segmentation process is as Figure 2 shown.

[0042] Correspondingly, the generated mapping relationship is as Figure 3 shown, where "Doc_01…Doc_06" represents the identification information corresponding to the alarm text data, "Server memory", "Severe", "Alarm" represent the alarm keywords, "√" indicates that the alarm keyword appears in the corresponding alarm text data, and "×" indicates that the alarm keyword does not appear in the corresponding alarm text data.

[0043] This method of dynamically adjusting the scanning window and combining multi-dictionary matching can more accurately extract meaningful keywords from the alarm text, providing a basis for subsequent information processing and analysis.

[0044] Such as Figure 4As shown, in an exemplary embodiment, before performing word segmentation on the alarm text data to obtain at least one alarm keyword, it further includes: obtaining a preset filtering rule, where the preset filtering rule includes filtering preset stop words and / or preset characters, and both the preset stop words and the preset characters are characters or words with a semantic contribution degree lower than a preset threshold; removing the keywords that meet the keyword filtering rule from the alarm keywords according to the preset filtering rule.

[0045] Specifically, the alarm text data is preprocessed through the preset filtering rule to remove the elements with relatively low semantic value therein, thereby optimizing the efficiency and accuracy of the subsequent word segmentation and processing process. Its essence is to identify and remove those characters or words that are common in natural language but contribute little to the core semantic expression of the text, enabling the subsequent processing to focus on the content with the ability to carry substantial information.

[0046] Specifically, the preset filtering rules are obtained, and these rules clarify the objects to be filtered, that is, the preset stop words and the preset characters. These stop words and characters are determined according to the semantic contribution degree lower than the preset threshold. In natural language, words such as "de", "le", "shi" play a certain role in the grammatical structure, but the core semantics they convey are very little and they are of little help in understanding the key information of the text.

[0047] Then, according to these preset filtering rules, the keywords that meet the rules are removed from the alarm text data. On the one hand, the amount of data to be processed during the subsequent word segmentation process is reduced, the computational burden is reduced, and the processing speed is increased. On the other hand, after removing these meaningless elements, the word segmentation result will more accurately reflect the core semantics of the text, avoiding interference from these redundant information to the subsequent analysis and retrieval, enabling the subsequent operations based on keywords to more accurately focus on the key information, and providing a more effective data basis for the entire alarm information processing process.

[0048] As Figure 4 shown, in an exemplary embodiment, before performing word segmentation on the alarm text data to obtain at least one alarm keyword, it further includes: determining whether there are sensitive keywords in the preset sensitive word library in the alarm keywords; if so, performing desensitization processing on the sensitive keywords, and the desensitization processing includes converting the sensitive keywords into corresponding standard keywords or encrypting the sensitive keywords.

[0049] Specifically, before the word segmentation process, sensitive word detection, desensitization, and special character conversion are performed on the alarm text data. The main purpose is to construct a secure and standardized layer for data preprocessing. By actively identifying and processing potential risk information and format interference factors, a secure and standardized data foundation is laid for subsequent word segmentation and the entire processing flow. The goal is to clean the original text before extracting key information, removing both sensitive content that may cause privacy leakage and format impurities that may lead to processing deviations, ensuring a balance between semantic integrity and security of the data.

[0050] Specifically, the sensitive word detection and desensitization mechanism actively scans and identifies privacy information or security-sensitive vocabulary (such as user identifiers, device authentication information, etc.) that may be contained in the alarm text data through a preset sensitive word library. Once a sensitive keyword is detected, desensitization processing is immediately performed, either by semantic mapping to convert it into a non-sensitive standard keyword (such as replacing a specific device number with a general device type), or by using encryption technology to perform irreversible processing on the sensitive content, cutting off the propagation path of sensitive information from the source. This preventive processing strategy not only meets the requirements of data compliance but also avoids the accidental leakage of sensitive information in subsequent links such as word segmentation, storage, and retrieval, ensuring that the alarm processing system meets information security specifications while operating efficiently.

[0051] For example Figure 4 , in an exemplary embodiment, the special character conversion mechanism maps format control characters, encoding conflict characters, or semantic ambiguity symbols (such as regular expression special symbols, non-standard encoding characters, etc.) existing in the alarm text data to recognizable standard formats through preset conversion rules. The core purpose of this process is to eliminate the interference of non-semantic-bearing characters on the subsequent processing flow. For example, to prevent special characters from causing misjudgment of word boundaries by the word segmentation algorithm or data storage anomalies during index construction. Through standardization processing, the text data only retains character units with actual semantics, ensuring that the word segmentation module can focus on extracting effective information and improving the accuracy of keyword extraction and the stability of the processing flow.

[0052] It can be seen that through the above methods, this embodiment solves the data privacy and security problems through sensitive word desensitization, and solves the data format and processing compatibility problems through special character conversion. The two work together to make the text data entering the word segmentation link meet both information security standards and have a high degree of semantic purity, providing a reliable input basis for subsequent efficient processing based on keywords (such as index construction, retrieval matching), and ensuring the security, robustness, and processing efficiency of the entire alarm information processing system from the bottom layer.

[0053] For example Figure 4, in one embodiment, a unified and non-redundant set of word segmentation results is constructed through the integration and deduplication mechanism of multi-source terms, ensuring that the semantic information of the alarm text is comprehensively and accurately extracted. Its essence is to systematically fuse the word segmentation results from different dictionary resources (such as general dictionaries, custom dictionaries, and professional dictionaries), eliminate duplicates while retaining all valid semantic units, and form a final list of terms that is both complete and accurate.

[0054] Specifically, since different dictionary resources are designed for general language, domain-specific terms, and user-defined vocabulary respectively, there may be overlapping (such as basic vocabulary included in both general and professional dictionaries) or complementary coverage (such as industry-specific terms unique to professional dictionaries and personalized vocabulary in custom dictionaries) in their word segmentation results. Through the merging operation, the term lists generated by each dictionary are combined at the set level to ensure that all possible valid terms (regardless of which dictionary they come from) are included in the candidate set, thus maximizing the retention of semantic information in the alarm text and avoiding the omission of keywords due to insufficient coverage of a single dictionary. The subsequent deduplication process then removes duplicate terms (such as repeated matches of the same word in different dictionaries) by identifying and eliminating them, eliminating the interference of redundant data on subsequent processing, and ensuring that each term exists uniquely in the final result and accurately reflects the text semantics.

[0055] This merging and deduplication mechanism ensures that the word segmentation results cover all language categories (general, professional, customized) involved in the text through the fusion of multi-dictionary word segmentation results, avoiding the loss of semantic information; through the deduplication operation, it ensures the simplicity of the term list, preventing repeated terms from causing data noise in subsequent steps such as index construction and similarity calculation, thus providing high-quality basic data input for the entire alarm information processing flow and fundamentally improving the reliability and processing efficiency of the word segmentation results.

[0056] For example Figure 5 , this application provides a method for processing alarm information, including:

[0057] S21: Obtain the target alarm keyword input by the user.

[0058] Specifically, an input interface for interacting with the user is constructed to obtain the target alarm keywords that the user is concerned about. Specifically, the user converts their query intention into recognizable text information by inputting the target alarm keywords. The target alarm keywords input by the user are based on their own focus and needs for alarm information, and may be words related to specific fault types, device names, or business scenarios. After obtaining the target alarm keywords, the goal and direction of subsequent processing are clarified, laying a foundation for accurately screening and positioning the required content in a large amount of alarm information. This input method guided by user needs can provide targeted alarm information query services according to the personalized needs of different users, improving the accuracy and effectiveness of information retrieval.

[0059] S22: Call the mapping relationship stored in the preset database. The mapping relationship stores the mapping relationship between the alarm keywords and the identification information of the alarm text data.

[0060] Specifically, utilize the mapping relationship stored in the preset database to build a bridge from the keywords to the identification information of the alarm text data. In the previous processing, the mapping relationship between the alarm keywords and the identification information of the alarm text data has been established and stored in the preset database. After obtaining the target alarm keywords input by the user, search for the mapping relationship related to the target keywords in this preset database. This mapping relationship is like an index table, which records the corresponding relationship between each alarm keyword and the identification information of the corresponding alarm text data.

[0061] By calling this mapping relationship, the identification information related to the target keywords can be quickly located without scanning all the alarm text data one by one, greatly improving the efficiency of information retrieval.

[0062] S23: Determine the target identification information according to the mapping relationship, and determine the target alarm text data according to the target identification information.

[0063] Specifically, based on the above-determined target identification information, achieve accurate positioning and acquisition from the identification information to the specific alarm text data. The target identification information is the unique identifier associated with the target alarm keywords, which points to the storage location of the corresponding alarm text data. According to these target identification information, the corresponding target alarm text data can be accurately found in the data source storing the alarm text data.

[0064] This step completes the transformation from the abstract target alarm keywords to the specific target alarm text information, enabling the user to obtain the detailed alarm content they are concerned about. Through this search method based on identification information, the required alarm information can be provided to the user efficiently and accurately, meeting the user's query and analysis needs for alarm information.

[0065] In an exemplary embodiment, determining the target identification information according to the mapping relationship includes: calculating the similarity between each alarm text data and the target alarm keyword; sorting the alarm text data from high to low according to the similarity, and determining the target identification information of the target alarm text data according to the sorting result.

[0066] Specifically, based on the existing mapping relationship, the target identification information is accurately determined through similarity calculation and sorting, so as to screen out the alarm text data that best meets the user's needs for the user. Its essence is to use a similarity evaluation mechanism to quantitatively compare numerous alarm text data with the target alarm keyword, and sort the data according to the user's needs for priority.

[0067] Specifically, calculate the similarity between each alarm text data and the target alarm keyword. This is a process of quantitative evaluation, which measures the matching degree of the alarm text and the keyword in terms of semantics, vocabulary, etc. through a specific algorithm. The calculation of similarity can reflect the relevance of each alarm text to the target keyword input by the user, providing an objective basis for subsequent sorting.

[0068] Then sort the alarm text data from high to low according to the similarity. The purpose of doing this is to rank the alarm text data most relevant to the target keyword at the top, facilitating the user to quickly obtain the most valuable information. According to the sorting result, determine the target identification information of the target alarm text data. These identification information point to the specific alarm text data in the storage system, and can accurately extract the alarm content most needed by the user, avoiding the user from searching for information in a large amount of irrelevant or low-relevance data, greatly improving the efficiency and accuracy of information retrieval, and meeting the user's need for accurately obtaining alarm information.

[0069] In an exemplary embodiment, calculating the similarity between each alarm text data and the target alarm keyword includes: setting a weight adjustment parameter for the target alarm keyword according to the user's needs; counting the number of occurrences of the target alarm keyword in each alarm text data, and calculating the word frequency parameter according to the number of occurrences; counting the total number of alarm text data containing the target alarm keyword, and calculating the inverse document frequency parameter according to the total number; calculating the similarity between each alarm text data and the target alarm keyword according to the weight adjustment parameter, the word frequency parameter and the inverse document frequency parameter.

[0070] Specifically, calculating the similarity between the alarm text data and the target alarm keyword combines the user's needs, the frequency of the keyword in the text, and the number of occurrences of the keyword in the overall document set to construct a comprehensive and objective similarity evaluation system.

[0071] First, set the weight adjustment parameter (boost) for the target alert keyword according to the user's needs. When different users query alert information, their attention levels to each alert keyword may vary. By setting the weight adjustment parameter, the importance that the user attaches to different keywords can be reflected, integrating the user's personalized needs into the similarity calculation. For example, for queries in certain key business scenarios, the user may assign a higher weight to specific keywords.

[0072] Next, count the number of occurrences of the target alert keyword in each alert text data, and calculate the term frequency parameter based on this (for example, the specific calculation formula is , where tf is the term frequency parameter, freq is the number of occurrences of the target alert keyword in the alert text data, k1 is the term protection parameter (default 1.2), b is the length normalization parameter (the impact of the word length on the entire document, default 0.75), dl is the length of the decomposed field in the current alert text data, and avgdl is the number of decomposed fields in the query alert text data / the number of query alert text data). The term frequency reflects how frequently the keyword appears in a single alert text. Generally speaking, the more times the keyword appears, the higher the correlation between this text and the keyword may be.

[0073] At the same time, count the total number of alert text data containing the target alert keyword, and then calculate the inverse document frequency parameter (for example, the specific calculation formula is , where idf is the inverse document frequency parameter, N is the total number of alert text data containing the query field, and n is the total number of alert text data containing the target alert keyword). The inverse document frequency measures the rarity of the keyword in the entire alert text dataset. If a keyword appears in fewer documents, then its inverse document frequency is higher, indicating that this keyword is more distinctive.

[0074] Finally, based on the weight adjustment parameter, the term frequency parameter, and the inverse document frequency parameter, comprehensively calculate the similarity between each alert text data and the target alert keyword (for example, the specific calculation formula is: similarity score = boost × tf × idf).

[0075] This comprehensive calculation method fully considers the user's needs, the local features of the keyword in the text, and the global features in the overall dataset, enabling the calculated similarity to more accurately reflect the association degree between the alert text and the target keyword, providing a reliable basis for subsequent sorting and screening, and helping users obtain the required alert information more precisely.

[0076] Generally speaking, the processing flow of alarm text data is divided into two parts: "storage" and "retrieval". In the "storage" part, the alarm text data is received, the data is segmented, the mapping relationship is constructed and stored in sequence; in the "retrieval" part, starting from obtaining the target alarm keyword, screening is carried out according to the mapping relationship, then similarity calculation is performed, and finally the results are sorted to form a complete alarm processing logic.

[0077] An embodiment of the present application also provides an electronic device, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any of the above embodiments of the processing method of alarm information.

[0078] For the introduction of the electronic device, please refer to the above embodiments, and the present application will not elaborate here.

[0079] An embodiment of the present application also provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps in any of the above embodiments of the processing method of alarm information when running.

[0080] In an exemplary embodiment, the above computer-readable storage medium may include, but is not limited to: various media such as USB flash drives, read-only memories (ROM for short), random access memories (RAM for short), mobile hard disks, magnetic disks or optical disks that can store computer programs.

[0081] For the introduction of the computer-readable storage medium, please refer to the above embodiments, and the present application will not elaborate here.

[0082] An embodiment of the present application also provides a computer program product. The above computer program product includes a computer program, and when the computer program is executed by a processor, it implements the steps in any of the above embodiments of the processing method of alarm information.

[0083] An embodiment of the present application also provides another computer program product, including a non-volatile computer-readable storage medium. The non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the steps in any of the above embodiments of the processing method of alarm information.

[0084] For the introduction of the computer program product, please refer to the above embodiments, and the present application will not elaborate here.

[0085] Those skilled in the art may further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered as exceeding the scope of this application.

[0086] The above has introduced in detail a method, device, and storage medium for processing warning information provided by this application. Specific examples have been used herein to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application. It should be noted that for those of ordinary skill in the art, without departing from the principle of this application, several improvements and modifications can be made to this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A method for processing warning information, characterized in that, Including: Receiving alarm text data; Performing word segmentation on the alarm text data to obtain at least one alarm keyword; Establishing a mapping relationship between the alarm keyword and the identification information of the alarm text data; Storing the mapping relationship in a preset database to retrieve corresponding target alarm text data from the preset database according to the target alarm keyword and the mapping relationship.

2. The method for processing the alarm information according to claim 1, wherein Performing word segmentation on the alarm text data to obtain at least one alarm keyword, including: Loading multiple dictionary resources and matching the alarm text data with the multiple dictionary resources; Performing word segmentation on the alarm text data according to the keywords in the multiple dictionary resources to obtain at least one alarm keyword.

3. The method for processing alarm information according to claim 2, wherein Performing word segmentation on the alarm text data according to the keywords in multiple dictionary resources to obtain at least one alarm keyword, including: Starting from the current starting position of the alarm text data, scanning character by character towards the end position, and initializing the scanning window to the preset maximum length; Matching the string within the scanning window with the multiple dictionary resources; If the match is successful, extracting the longest string matched by the current scanning window as the alarm keyword for word segmentation processing, and moving the scanning pointer to the next character position after the end of the longest string; If the match fails, gradually reducing the scanning window and determining whether the number of characters in the scanning window is greater than 1. If it is greater than 1, then return to the step of matching the string within the scanning window with the multiple dictionary resources to perform a re-match until the match is successful or until the number of characters in the scanning window is equal to 1. Taking the string or single character that matches successfully within the scanning window as the alarm keyword for word segmentation processing, and moving the scanning pointer to the next character position; Updating the next character position to the new current starting position, and returning to the step of starting from the current starting position of the alarm text data, scanning character by character towards the end position, and initializing the scanning window to the preset maximum length according to the new starting position to continue matching the remaining alarm text data except for the single character until the word segmentation processing of all characters in the alarm text data is completed.

4. The method for processing alarm information according to any one of claims 1-3, characterized in that Before performing word segmentation on the alarm text data to obtain at least one alarm keyword, it further includes: Obtaining a preset filtering rule, where the preset filtering rule includes filtering preset stop words and / or preset characters, and both the preset stop words and the preset characters are characters or words with a semantic contribution degree lower than a preset threshold; Removing the keywords that meet the keyword filtering rule from the alarm keywords according to the preset filtering rule.

5. The method for processing alarm information according to any one of claims 1-3, characterized in that Before performing word segmentation on the alarm text data to obtain at least one alarm keyword, it further includes: Judging whether the alarm keywords contain sensitive keywords in a preset sensitive word library; If it contains, performing desensitization processing on the sensitive keyword, where the desensitization processing includes converting the sensitive keyword into a corresponding standard keyword or encrypting the sensitive keyword.

6. A method for processing warning information, characterized in that, Including: Obtaining the target alarm keyword input by the user; Call the mapping relationship stored in the preset database, where the mapping relationship stores the mapping relationship between the alarm keyword and the identification information of the alarm text data; Determine the target identification information according to the mapping relationship, and determine the target alarm text data according to the target identification information.

7. The method for processing warning information according to claim 6, characterized in that, Determining the target identification information according to the mapping relationship includes: Calculate the similarity between each of the alarm text data and the target alarm keyword; Sort the alarm text data in descending order of similarity, and determine the target identification information of the target alarm text data according to the sorting result.

8. The method for processing alarm information according to claim 7, wherein Calculating the similarity between each of the alarm text data and the target alarm keyword includes: Set a weight adjustment parameter for the target alarm keyword according to user requirements; Count the number of occurrences of the target alarm keyword in each of the alarm text data, and calculate the term frequency parameter according to the number of occurrences; Count the total number of alarm text data containing the target alarm keyword, and calculate the inverse document frequency parameter according to the total number; Calculate the similarity between each of the alarm text data and the target alarm keyword according to the weight adjustment parameter, the term frequency parameter and the inverse document frequency parameter.

9. An electronic device, characterized in that, Includes: A memory for storing a computer program; A processor for implementing the steps of the alarm information processing method according to any one of claims 1 to 8 when executing the computer program.

10. A computer-readable storage medium, characterized in that, A computer program is stored in the computer-readable storage medium, wherein the computer program implements the steps of the alarm information processing method according to any one of claims 1-8 when executed by a processor.

Images