Universal text content analysis and reasoning system based on artificial intelligence

Abstract

A universal text content analysis and reasoning system based on artificial intelligence comprises a review engine unit, a text analysis unit, a rule analysis unit, a reasoning unit and a thread management unit, the review engine unit establishes a preprocessing process and a rule acquisition process, and when a review task is executed, the preprocessing process and the rule acquisition process are executed; the method comprises the following steps of: quantifying a task load at a previous moment or a previous period to obtain a quantitative mark, and allocating resource occupation of a preprocessing process and a rule acquisition process according to the quantitative mark; the text analysis unit preprocesses a to-be-processed text to obtain a data signal; the rule analysis unit is used for crawling the basic rule and carrying out semantic analysis on the basic rule to obtain an inference rule; the reasoning unit inputs the data signal and a vector corresponding to the reasoning rule into a reasoning neural network for logic calculation to obtain a reasoning result; and the thread management unit manages threads in the preprocessing process and the rule acquisition process. According to the invention, the resource allocation is reasonable, and the task execution speed is improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology. More specifically, it particularly relates to a general text content analysis and reasoning system based on artificial intelligence. Background Art

[0002] With the continuous update of laws, regulations and industry standards, enterprises need to ensure that the legal and regulatory documents involved in their operations comply with the requirements of relevant laws, regulations and / or internal policies during the operation process. For example, whether the contract complies with the requirements of relevant laws, regulations and internal policies; whether the formulated processes comply with the requirements of internal policies; whether the formulated rules and regulations comply with relevant laws and regulations.

[0003] The traditional compliance review of legal and regulatory documents mainly relies on manual work. Reviewers need to consult relevant laws, regulations or internal policy documents and compare them with each other, which is inefficient, costly and error-prone, and it is difficult to cope with complex legal environments and rapidly changing compliance requirements.

[0004] With the development of information technology, although there are some simple text retrieval tools to assist in the review, they cannot achieve in-depth semantic analysis and intelligent judgment, and it is difficult to meet the growing demand for institutional compliance review.

[0005] Therefore, there is an urgent need for an intelligent text content analysis and reasoning assistance system. Summary of the Invention

[0006] (1) Invention Objective: To solve the problems existing in the above-mentioned prior art, the objective of the present invention is to provide a general text content analysis and reasoning system based on artificial intelligence, which can automatically identify, analyze and evaluate the compliance of institutional texts and provide improvement suggestions.

[0007] (2) Technical Solution: To solve the above technical problems, the present technical solution provides a general text content analysis and reasoning system based on artificial intelligence, including a review engine unit, a text analysis unit, a rule analysis unit, an inference unit and a thread management unit. The review engine unit allocates the resource occupation of each unit when performing a review task and establishes a preprocessing process and a rule acquisition process; the text analysis unit is used to preprocess the text to be processed to obtain a data signal; the rule analysis unit is used to crawl basic rules and perform semantic analysis on the basic rules to obtain inference rules; the inference unit inputs the data signal and the vector corresponding to the inference rule into an inference neural network for logical calculation to obtain an inference result; the thread management unit manages the threads in the preprocessing process and the rule acquisition process respectively. When the review engine unit executes a review task, it quantifies the task volume at the previous moment or the previous cycle. After obtaining a quantization mark, it allocates the resource occupancy of the preprocessing process and the rule acquisition process according to the quantization mark.

[0008] In the general artificial intelligence-based text content analysis and reasoning system, preprocessing the text to be processed includes performing semantic analysis on the text to be processed and extracting key data of the text; after organizing the key data, converting the format to obtain the data signal.

[0009] In the general artificial intelligence-based text content analysis and reasoning system, the rule analysis unit converts each inference rule into a vector according to a domain-specific language. Each inference rule corresponds to a vector, and all the vectors form a vector space of the rules.

[0010] In the general artificial intelligence-based text content analysis and reasoning system, the inference unit constructs a knowledge graph according to the association relationship between entities in the inference results of the same text to be processed.

[0011] In the general artificial intelligence-based text content analysis and reasoning system, the display unit converts the knowledge graph into a target report through a visualization engine according to the input unit and displays the target report.

[0012] In the general artificial intelligence-based text content analysis and reasoning system, the review engine unit allocates the resource occupancy of the preprocessing process and the rule acquisition process, specifically including: The review engine unit obtains the task volume to be processed at the previous moment or the previous cycle; The review engine unit quantifies the task volume to be processed at the previous moment or the previous cycle to obtain a quantization mark; The review engine unit calculates the number of quantization marks at the previous moment or the previous cycle to obtain a quantization value; The review engine unit calculates the resource occupancy ratio of the rule analysis unit and the rule analysis unit at the previous moment or the previous cycle according to the quantization value at the previous moment or the previous cycle; The review engine unit allocates the resource occupancy of the rule analysis unit and the rule analysis unit at the current moment or the current cycle according to the resource occupancy ratio of the rule analysis unit and the rule analysis unit at the previous moment or the previous cycle.

[0013] The described general artificial intelligence-based text content analysis and reasoning system, wherein the review engine unit quantifies the amount of tasks to be processed at the previous moment or in the previous cycle to obtain a quantization mark. Specifically, the review engine unit quantifies the number of texts to be processed and the memory occupancy at the previous moment or in the previous cycle to obtain a quantization mark for the texts to be processed; the review engine unit quantifies the number of basic rules to be processed and the memory occupancy at the previous moment or in the previous cycle to obtain a quantization mark for the basic rules.

[0014] The described general artificial intelligence-based text content analysis and reasoning system, wherein the review engine unit calculates the number of quantization marks at the previous moment or in the previous cycle to obtain a quantization value. Specifically, the review engine unit calculates the number of quantization marks for the texts to be processed at the previous moment or in the previous cycle to obtain a first quantization value; the review engine unit calculates the number of quantization marks for the basic rules to be processed at the previous moment or in the previous cycle to obtain a second quantization value.

[0015] The described general artificial intelligence-based text content analysis and reasoning system, wherein the review engine unit quantifies the amount of tasks to be processed at the previous moment or in the previous cycle to obtain a quantization mark, specifically including that the review engine unit generates the same number of first quantization marks according to the number of tasks to be processed at the previous moment or in the previous cycle; the review engine unit divides the size of each task to be processed at the previous moment or in the previous cycle by a first quantization threshold to obtain a quantization quotient value; the review engine unit generates a target second quantization mark according to the quantization quotient value; the first quantization mark and the second quantization mark form the quantization mark.

[0016] The described general artificial intelligence-based text content analysis and reasoning system, wherein when the quantization quotient value of each task is less than 1, the number of second quantization marks is zero, and at this time, the second quantization mark is zero. When there is a quantization quotient value greater than 1 among the quantization quotient values of all tasks, the number of the second quantization marks is greater than zero: when the quantization quotient value of a task is an integer greater than 1, the review engine unit generates the number of second quantization marks equal to the quantization quotient value of the task minus 1, which is the second quantization mark of the task; when the quantization quotient value of a task is a non-integer greater than 1, the review engine unit generates the number of second quantization marks equal to the quantization quotient value of the task, which is the second quantization mark of the task. The second quantization marks of all tasks form the target second quantization mark.

[0017] (3) Beneficial effects: The present invention provides a general artificial intelligence-based text content analysis and inference system, which executes the preprocessing task and rule acquisition task of the text to be processed through two relatively independent processes, and allocates the resource occupancy at the current moment or current cycle according to the task volume of the preprocessing task and rule acquisition task at the previous moment or previous cycle, making the resource allocation for the preprocessing process and rule acquisition process more reasonable and improving the task execution speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 FIG. is a schematic diagram of the steps when the review engine unit of a general artificial intelligence-based text content analysis and inference system of the present invention allocates the resource occupancy of the preprocessing process and rule acquisition process; Figure 2 FIG. is a schematic diagram of the partition of the coordinates to be allocated in a general artificial intelligence-based text content analysis and inference system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The present invention will be further described in detail below in conjunction with the preferred embodiments. More details are set forth in the following description to facilitate a full understanding of the present invention. However, the present invention is clearly capable of being implemented in many other ways different from this description. Those skilled in the art can make similar generalizations and deductions according to the actual application situation without departing from the connotation of the present invention. Therefore, the protection scope of the present invention should not be limited by the content of this specific embodiment.

[0020] The accompanying drawings are schematic diagrams of the embodiments of the present invention. It should be noted that this accompanying drawing is only an example and is not drawn according to the condition of equal proportion, and should not be used to limit the actual protection scope required by the present invention.

[0021] A general artificial intelligence-based text content analysis and inference system is used to analyze and infer the text to be processed, realize the review of the compliance of the text to be processed, and complete the review task. A general artificial intelligence-based text content analysis and inference system includes a review engine unit, a text analysis unit, a rule analysis unit, an inference unit, a display unit, and a thread management unit.

[0022] The review engine unit is deployed on the underlying operation platform and performs core scheduling for the system. Specifically, the review engine unit allocates the resource occupancy of each unit when executing the review task. Each unit includes a text analysis unit, a rule analysis unit, an inference unit, and a display unit. The underlying operation platform includes a desktop operating system and a mobile operating system.

[0023] The text analysis unit performs preprocessing tasks on the text to be processed, that is, the text analysis unit is used to preprocess the text to be processed to obtain a data signal. Preprocessing the text to be processed includes performing semantic analysis on the text to be processed and extracting the key data of the text; after organizing the key data, converting the format to obtain the data signal. The key data refers to the variable data in the text to be processed. For example, when the text to be processed is a contract, the key data refers to the specific terms or specific numerical values or terms in the terms; when the text to be processed is business data, the key data refers to the core data in the specific business data. The key data can be obtained by performing semantic analysis on the text to be processed or directly obtained from the text to be processed, and no specific display is made here. The data signal is the variable information required in the inference unit, and its format is the format recognizable when the inference unit performs logical calculations. Performing semantic analysis on the text to be processed can be done through artificial intelligence for semantic analysis, such as DeepSeek. Organizing the key data includes preliminary processing such as calculating and sorting the key data.

[0024] The rule analysis unit performs rule acquisition tasks, that is, the rule analysis unit is used to crawl basic rules and perform semantic analysis on the basic rules to obtain inference rules. The basic rules include legal regulations or industry rules such as text logic rules, calculation rules, and business analysis rules. The rule analysis unit performs semantic analysis on the basic rules to obtain inference rules, and specifically, it can perform semantic analysis through artificial intelligence, such as DeepSeek.

[0025] The inference rules can also be input through the input unit.

[0026] The rule analysis unit stores the inference rules in the rule library, and the rule library includes inference rules for different industries and different classifications. When the rule analysis unit performs semantic analysis on the basic rules, it compares the obtained inference rules with the inferences in the rule library and replaces the inference rules in the rule library that conflict with or are incomplete with them, so as to ensure that the inference rules in the rule library comply with the latest laws and regulations.

[0027] The rule analysis unit can also convert each inference rule into a vector according to DSL (Domain-Specific Language). Each inference rule corresponds to a vector, and all vectors form the vector space of the rules.

[0028] The inference unit inputs the data signal and the vector corresponding to the inference rule into the inference neural network for logical calculation to obtain an inference result. The inference neural network uses the vector corresponding to the inference rule as a calculation function and calculates with the data signal as the independent variable, and the calculation result obtained is the inference result.

[0029] The inference unit can also construct a knowledge graph from the inference results of the same text to be processed. The inference unit constructs the knowledge graph based on the association relationships between entities in the inference results of the same text to be processed.

[0030] The display unit converts the knowledge graph into a target report through a visualization engine according to the format of the output report selected by the input unit, and displays the target report. The target report includes multiple formats, which can be text format, presentation format, chart format, etc.

[0031] The review engine unit supports calling system APIs. The review engine unit allocates system resources for the preprocessing tasks of the text analysis unit and the rule acquisition tasks of the rule analysis unit, and establishes a process corresponding to the preprocessing tasks of the text analysis unit: the preprocessing process, and a process corresponding to the rule tasks of the rule analysis unit: the rule acquisition process. The preprocessing process and the rule acquisition process are two independent processes in a general artificial intelligence-based text content analysis and reasoning system.

[0032] The preprocessing process executes the preprocessing of multiple texts to be processed by the text analysis unit at the same time. The preprocessing of each text to be processed is executed by different threads in the preprocessing process, so as to realize the preprocessing of each text to be processed separately.

[0033] The rule acquisition process executes the semantic analysis of multiple basic rules by the rule analysis unit at the same time. The semantic analysis of each basic rule is executed by different threads in the rule acquisition process, so as to realize the semantic analysis of each basic rule separately.

[0034] The thread management unit manages the threads in the preprocessing process and the rule acquisition process respectively, including: the occupation and allocation of the current process resources by each thread in the preprocessing process and the rule acquisition process.

[0035] The thread management unit allocates the resource occupation of each thread in the process according to the number of tasks to be processed in the process. Specifically, the thread management unit manages each thread in the preprocessing process according to the number of texts to be processed by the text analysis unit, and allocates the use of the current process resources by each thread; the thread management unit manages each thread in the rule acquisition process according to the number of basic rules to be processed by the rule analysis unit, and allocates the use of the current process resources by each thread.

[0036] The thread management unit allocates the resource occupation of each thread in the process, specifically, it can allocate the occupation of CPU or kernel resources.

[0037] When the review engine unit performs a review task, it allocates resources for the rule analysis unit and the resource occupancy of the rule analysis unit, that is, it allocates resources for the preprocessing process and the rule acquisition process. When the review engine unit allocates CPU or kernel resource occupancy for the preprocessing process and the rule acquisition process, after quantifying the amount of tasks to be processed in different time periods, it allocates resource occupancy according to the quantification of the task amount. Specifically, it quantifies the task amount at the previous moment or the previous cycle, obtains a quantification mark, and then allocates resource occupancy according to the quantification mark. The same moment in the previous cycle refers to the previous cycle adjacent to the cycle where the current moment is located.

[0038] As Figure 1 shown, the specific implementation process is as follows: The review engine unit obtains the amount of tasks to be processed at the previous moment or the previous cycle. Specifically, the review engine unit obtains the number of texts to be processed and the size of each text to be processed at the previous moment or the previous cycle; the review engine unit obtains the number of basic rules to be processed and the size of each basic rule to be processed at the previous moment or the previous cycle. The size of the text to be processed refers to the storage space capacity occupied by the text to be processed, and the size of the basic rule to be processed refers to the storage space capacity occupied by the basic rule to be processed.

[0039] The review engine unit quantifies the amount of tasks to be processed at the previous moment or the previous cycle to obtain a quantification mark. Specifically, the review engine unit quantifies the number of texts to be processed and the memory occupancy at the previous moment or the previous cycle to obtain a quantification mark for the text to be processed; the review engine unit quantifies the number of basic rules to be processed and the memory occupancy at the previous moment or the previous cycle to obtain a quantification mark for the basic rule.

[0040] The review engine unit calculates the number of quantification marks at the previous moment or the previous cycle to obtain a quantification value. Specifically, the review engine unit calculates the number of quantification marks for the texts to be processed at the previous moment or the previous cycle to obtain a first quantification value; the review engine unit calculates the number of quantification marks for the basic rules to be processed at the previous moment or the previous cycle to obtain a second quantification value.

[0041] The review engine unit calculates the resource occupancy ratio of the rule analysis unit and the rule analysis unit at the previous moment or the previous cycle according to the quantification value at the previous moment or the previous cycle. Specifically, the review engine unit calculates the ratio of the first quantification value and the second quantification value, which is the resource occupancy ratio of the rule analysis unit and the rule analysis unit.

[0042] The review engine unit allocates the resource occupancy of the rule analysis unit and the rule analysis unit at the previous moment or previous cycle to the resource occupancy of the rule analysis unit and the rule analysis unit at the current moment or current cycle according to the rule analysis unit and the resource occupancy ratio of the rule analysis unit at the previous moment or previous cycle.

[0043] The review engine unit includes a first quantization threshold.

[0044] The review engine unit quantifies the amount of tasks to be processed at the previous moment or previous cycle to obtain a quantization mark. Specifically, the review engine unit generates the same number of first quantization marks according to the number of tasks to be processed at the previous moment or previous cycle; the review engine unit divides the size of each task to be processed at the previous moment or previous cycle by the first quantization threshold to obtain a quantization quotient value; the review engine unit generates a target second quantization mark according to the quantization quotient value; the first quantization mark and the second quantization mark form the quantization mark.

[0045] When the quantization quotient value of each task is less than 1, the number of second quantization marks is zero, and at this time the second quantization mark is zero. When there is a quantization quotient value greater than 1 among the quantization quotient values of all tasks, the number of the second quantization marks is greater than zero: when the quantization quotient value of the task is an integer greater than 1, the review engine unit generates the second quantization mark with the number of the quantization quotient value of the task minus 1, that is, the second quantization mark of the task; when the quantization quotient value of the task is a non-integer greater than 1, the review engine unit generates the second quantization mark with the number of the quantization quotient value of the task, that is, the second quantization mark of the task. The second quantization marks of all tasks form the target second quantization mark.

[0046] The quantization mark can be a bar chart or other images representing quantities, and no specific limitation is made here. The number of the first quantization marks is the same as the number of tasks to be processed at the previous moment or previous cycle. Dividing the size of the tasks to be processed at the previous moment or previous cycle by the first quantization threshold means taking the size of each task to be processed at the previous moment or previous cycle as the dividend and the first quantization threshold as the divisor for division operation. The size of each task refers to the memory occupancy of each task.

[0047] Specifically, the review engine unit quantifies the number and memory occupancy of the text to be processed at the previous moment or previous cycle to obtain the quantization mark of the text to be processed; the review engine unit quantifies the number and memory occupancy of the basic rules to be processed at the previous moment or previous cycle to obtain the quantization mark of the basic rules.

[0048] The review engine unit generates the same number of first text quantization tags according to the number of texts to be processed at the previous moment or the previous cycle; the review engine unit divides the size of each text to be processed at the previous moment or the previous cycle by the first quantization threshold to obtain a text quantization quotient value; the review engine unit generates a second quantization tag for the target text according to the text quantization quotient value; the first text quantization tag and the second quantization tag for the target text form the quantization tag of the text to be processed.

[0049] When the text quantization quotient value of each text to be processed is less than 1, the number of second text quantization tags is zero, and at this time the second text quantization tag is zero. When there is a text quantization quotient value greater than 1 among the text quantization quotient values of all texts to be processed, the number of the second text quantization tags is greater than zero: when the text quantization quotient value of the text to be processed is an integer greater than 1, the review engine unit generates the second text quantization tag with the number being the text quantization quotient value of the text to be processed minus 1, which is the second text quantization tag of the text to be processed; when the text quantization quotient value of the text to be processed is a non-integer greater than 1, the review engine unit generates the second text quantization tag with the number being the text quantization quotient value of the text to be processed, which is the second text quantization tag of the text to be processed. The second text quantization tags of all texts to be processed form the second quantization tag for the target text.

[0050] The review engine unit generates the same number of first rule quantization tags according to the number of basic rules to be processed at the previous moment or the previous cycle; the review engine unit divides the size of each basic rule to be processed at the previous moment or the previous cycle by the first quantization threshold to obtain a rule quantization quotient value; the review engine unit generates a second quantization tag for the target rule according to the rule quantization quotient value; the first rule quantization tag and the second quantization tag for the target rule form the quantization tag of the basic rule.

[0051] When the rule quantization quotient value of each basic rule to be processed is less than 1, the number of second rule quantization tags is zero, and at this time the second rule quantization tag is zero. When there is a rule quantization quotient value greater than 1 among the rule quantization quotient values of all basic rules to be processed, the number of the second rule quantization tags is greater than zero: when the rule quantization quotient value of the basic rule to be processed is an integer greater than 1, the review engine unit generates the second rule quantization tag with the number being the rule quantization quotient value of the basic rule to be processed minus 1, which is the second rule quantization tag of the basic rule to be processed; when the rule quantization quotient value of the basic rule to be processed is a non-integer greater than 1, the review engine unit generates the second rule quantization tag with the number being the rule quantization quotient value of the basic rule to be processed, which is the second rule quantization tag of the basic rule to be processed. The second rule quantization tags of all basic rules to be processed form the second quantization tag for the target rule.

[0052] When the thread management unit allocates the resource occupation of each thread in a process, it allocates the resource occupation of each thread in the preprocessing process and the rule acquisition process. The resource occupation of each thread refers to the allocation of thread running resources, including but not limited to registers, stack space, etc.

[0053] Specifically, it includes: The thread management unit divides a cache pool for each process. The cache pool is used to store the threads in the corresponding process, and each thread corresponds to a preprocessing task or a rule acquisition task.

[0054] The thread management unit creates threads for the tasks in the cache pool of each process, and allocates running resources to the created threads. After each thread obtains the running resources, it executes the thread task.

[0055] After the thread completes the task, the thread management unit destroys the thread and reclaims the running resources.

[0056] Each thread in the cache pool includes corresponding task data. The task data refers to the time when the task corresponding to the thread enters the cache pool, the memory occupancy of the task file, etc. When the thread management unit allocates running resources to the created threads, it allocates running resources to the created threads through coordinate partitioning. Specifically, it includes the following steps: Establish a thread resource allocation coordinate. The thread resource allocation coordinate system is a two-dimensional coordinate. The coordinate axes of the thread resource allocation coordinate respectively represent the time when the task corresponding to the thread enters the cache pool and the memory occupancy of the task file. Here, taking the x-axis to represent the time when the task corresponding to the thread enters the cache pool and the y-axis to represent the memory occupancy of the task file as an example for explanation.

[0057] According to the task data corresponding to each thread, mark the created threads in the cache pool in the thread resource coordinate system to obtain an allocation coordinate system to be allocated. Each thread corresponds to a coordinate point in the thread resource coordinate system. The allocation coordinate system includes the thread resource coordinate system and the coordinate points corresponding to all the threads of the process.

[0058] Perform hierarchical partitioning on the allocation coordinate system to be allocated. All partitions include all the coordinate points, or in other words, the area composed of all partitions covers all the coordinate points in the allocation coordinate system to be allocated. The hierarchical partitioning includes at least three levels. Here, taking three levels of partitioning as an example, it specifically includes a first partition, a second partition, and a third partition. The area composed of the first partition, the second partition, and the third partition covers all the coordinate points in the allocation coordinate system to be allocated.

[0059] Allocate running resources to the threads corresponding to the coordinate points according to the partitions where the coordinate points are located.

[0060] As Figure 2 shown, the first partition is a "「" - shaped partition composed of a 1 - 1 partition and a 1 - 2 partition. The range of the 1 - 1 partition on the x - axis is greater than or equal to the minimum entry time of all tasks entering the cache pool and less than or equal to the maximum entry time of all tasks entering the cache pool; the range of the 1 - 1 partition on the y - axis is less than or equal to the maximum memory occupancy of all task files and greater than or equal to the first memory threshold. The range of the 1 - 2 partition on the x - axis is greater than or equal to the minimum entry time of all tasks entering the cache pool and less than or equal to the first time threshold; the range of the 1 - 2 partition on the y - axis is less than or equal to the first memory threshold and greater than or equal to the minimum memory occupancy of all task files.

[0061] The second partition is a "「" - shaped partition composed of a 2 - 1 partition and a 2 - 2 partition. The range of the 2 - 1 partition on the x - axis is greater than the first time threshold and less than or equal to the maximum entry time of all tasks entering the cache pool; the range of the 2 - 1 partition on the y - axis is less than the first memory threshold and greater than or equal to the second memory threshold. The range of the 2 - 2 partition on the x - axis is greater than the first time threshold and less than or equal to the second time threshold; the range of the 2 - 2 partition on the y - axis is less than the first memory threshold and greater than or equal to the minimum memory occupancy of all task files.

[0062] The third partition is a "|" - shaped partition. The range of the third partition on the x - axis is greater than the second time threshold and less than or equal to the maximum entry time of all tasks entering the cache pool; the range of the third partition on the y - axis is less than the second memory threshold and greater than or equal to the maximum memory occupancy of all task files.

[0063] The difference between the maximum memory occupancy and the first memory threshold, the difference between the first memory threshold and the second memory threshold, and the difference between the second memory threshold and the minimum memory occupancy are respectively equal.

[0064] The difference between the minimum entry time and the first time threshold, the difference between the first time threshold and the second time threshold, and the difference between the second time threshold and the maximum entry time are respectively equal.

[0065] When allocating running resources for the thread corresponding to the coordinate point according to the partition where the coordinate point is located, according to the running resources required by the thread corresponding to the coordinate point, the running resources are allocated in the order of the first partition, the second partition, and the third partition.

[0066] Specifically, Calculate the required running resources for the threads corresponding to the coordinate points in the first partition, second partition, and third partition respectively. When the sum of the required running resources for the threads corresponding to the coordinate points in the first partition, second partition, and third partition is less than or equal to the running resources of the current process, the thread management unit simultaneously allocates running resources to the threads corresponding to the coordinate points in the first partition, second partition, and third partition.

[0067] When the sum of the required running resources for the threads corresponding to the coordinate points in the first partition, second partition, and third partition is greater than the running resources of the current process: Calculate the sum of the required running resources for the threads corresponding to the coordinate points in the first partition and second partition. When the sum of the required running resources for the threads corresponding to the coordinate points in the first partition and second partition is less than or equal to the running resources of the current process, the thread management unit simultaneously allocates running resources to the threads corresponding to the coordinate points in the first partition and second partition; after the thread management unit completes the allocation of running resources for the threads corresponding to the coordinate points in the first partition and second partition, it allocates running resources to the threads corresponding to the coordinate points in the third partition in descending order of the y - coordinate of the coordinate points until the remaining running resources of the process are less than the required running resources of all the threads corresponding to the coordinate points.

[0068] When the sum of the required running resources for the threads corresponding to the coordinate points in the first partition and second partition is greater than the running resources of the current process: Calculate the sum of the required running resources for the threads corresponding to the coordinate points in the first partition. When the sum of the required running resources for the threads corresponding to the coordinate points in the first partition is less than or equal to the running resources of the current process, the thread management unit simultaneously allocates running resources to the threads corresponding to the coordinate points in the first partition; after the thread management unit completes the allocation of running resources for the threads corresponding to the coordinate points in the first partition, it allocates running resources to the threads corresponding to the coordinate points in the second partition and third partition in descending order of the y - coordinate of the coordinate points and from the second partition to the third partition until the remaining running resources of the process are less than the required running resources of all the threads corresponding to the coordinate points.

[0069] When the sum of the required running resources for the threads corresponding to the coordinate points in the first partition is greater than the running resources of the current process: The thread management unit allocates running resources to the threads corresponding to the coordinate points in the first partition, second partition, and third partition in descending order of the y - coordinate of the coordinate points, ascending order of the x - coordinate of the coordinate points, from the first partition to the second partition, and from the second partition to the third partition until the remaining running resources of the process are less than the required running resources of all the threads corresponding to the coordinate points.

[0070] By allocating running resources to the threads in each process in the above manner, when ensuring that the running resources of the threads are sufficient to guarantee thread operation, the running resources of the process are planned, the execution speed of the threads is increased, and the running speed of the system is greatly improved. At the same time, by allocating the running resources of the threads according to the situation of the threads, the running progress of all threads is ensured.

[0071] In a general artificial intelligence-based text content analysis and inference system, it further includes a storage unit, and the storage unit is used to store the data pre-existing in the system, as well as task data such as process data and result data in the review task.

[0072] When the review engine unit establishes the preprocessing process and the rule acquisition process, a shared storage area is set in the storage unit, and the shared storage area stores the data signal of the text to be processed, the rule library, and / or the vector space. The preprocessing process and the rule acquisition process include the call permission for the shared storage area. The text analysis unit stores the data signal in the shared storage area, and the rule analysis unit stores the rule library and / or the vector space in the shared storage area.

[0073] A general artificial intelligence-based text content analysis and inference system executes the preprocessing task and the rule acquisition task of the text to be processed through two relatively independent processes, which improves the system task execution speed. Moreover, when allocating resources to the two processes, according to the task amounts of the preprocessing task and the rule acquisition task in the previous moment or previous cycle, the resource occupation in the current moment or current cycle is allocated, making the resource allocation for the preprocessing process and the rule acquisition process more reasonable, improving the task execution speeds of the preprocessing process and the rule acquisition process, and further ensuring the system task execution speed.

[0074] The above content is an illustration of the preferred embodiments of the present invention, which can help those skilled in the art to more fully understand the technical solution of the present invention. However, these embodiments are merely examples and cannot be considered that the specific implementation manners of the present invention are limited to the descriptions of these embodiments. For those of ordinary skill in the technical field to which the present invention belongs, without departing from the concept of the present invention, several simple deductions and transformations can still be made, and all should be regarded as belonging to the protection scope of the present invention.

Claims

1. A general artificial intelligence-based text content analysis and reasoning system, characterized in that, It includes a review engine unit, a text analysis unit, a rule analysis unit, an inference unit, and a thread management unit. The review engine unit allocates the resource occupancy of each unit during the execution of the review task and establishes a preprocessing process and a rule acquisition process; the text analysis unit is used to preprocess the text to be processed to obtain a data signal; the rule analysis unit is used to crawl basic rules and perform semantic analysis on the basic rules to obtain inference rules; the inference unit inputs the vectors corresponding to the data signal and the inference rules into an inference neural network for logical calculation to obtain an inference result; the thread management unit manages the threads in the preprocessing process and the rule acquisition process respectively. When the review engine unit executes the review task, after quantifying the task volume of the previous moment or the previous cycle to obtain a quantization mark, it allocates the resource occupancy of the preprocessing process and the rule acquisition process according to the quantization mark.

2. The general artificial intelligence-based text content analysis and inference system according to claim 1, characterized in that Preprocessing the text to be processed includes performing semantic analysis on the text to be processed to extract the key data of the text; after organizing the key data, converting the format to obtain the data signal.

3. A general artificial intelligence-based text content analysis and reasoning system according to claim 1, characterized in that, The rule analysis unit converts each inference rule into a vector according to a domain-specific language. Each inference rule corresponds to a vector, and all vectors form a vector space of the rules.

4. A general artificial intelligence-based text content analysis and reasoning system according to claim 1, characterized in that, The inference unit constructs a knowledge graph according to the association relationship between entities in the inference results of the same text to be processed.

5. A general artificial intelligence-based text content analysis and reasoning system according to claim 1, characterized in that, It further includes a display unit. The display unit converts the knowledge graph into a target report through a visualization engine according to the input unit and displays the target report.

6. The general artificial intelligence-based text content analysis and reasoning system according to claim 1, characterized in that, The review engine unit allocates the resource occupancy of the preprocessing process and the rule acquisition process, specifically including The review engine unit obtains the task volume to be processed at the previous moment or the previous cycle. The review engine unit quantifies the task volume to be processed at the previous moment or the previous cycle to obtain a quantization mark. The review engine unit calculates the quantity of the quantization mark at the previous moment or the previous cycle to obtain a quantization value. The review engine unit calculates the resource occupancy ratio of the rule analysis unit and the rule analysis unit at the previous moment or the previous cycle according to the quantization value at the previous moment or the previous cycle. The review engine unit allocates the resource occupancy of the rule analysis unit and the rule analysis unit at the current moment or the current cycle according to the resource occupancy ratio of the rule analysis unit and the rule analysis unit at the previous moment or the previous cycle.

7. A general artificial intelligence-based text content analysis and reasoning system according to claim 6, characterized in that, The review engine unit quantifies the task volume to be processed at the previous moment or the previous cycle to obtain a quantization mark. Specifically, the review engine unit quantifies the quantity and memory occupancy of the text to be processed at the previous moment or the previous cycle to obtain a quantization mark of the text to be processed. The review engine unit quantifies the quantity and memory occupancy of the basic rules to be processed at the previous moment or the previous cycle to obtain a quantization mark of the basic rules.

8. A general artificial intelligence-based text content analysis and reasoning system according to claim 7, characterized in that, The review engine unit calculates the number of quantization tags at the previous moment or the previous cycle to obtain a quantization value. Specifically, the review engine unit calculates the number of quantization tags of the text to be processed at the previous moment or the previous cycle to obtain a first quantization value; the review engine unit calculates the number of quantization tags of the basic rules to be processed at the previous moment or the previous cycle to obtain a second quantization value.

9. The general artificial intelligence-based text content analysis and reasoning system according to claim 6, characterized in that, The review engine unit quantizes the amount of tasks to be processed at the previous moment or the previous cycle to obtain quantization tags, specifically including that the review engine unit generates the same number of first quantization tags according to the number of tasks to be processed at the previous moment or the previous cycle; The review engine unit divides the size of each task to be processed at the previous moment or the previous cycle by a first quantization threshold to obtain a quantization quotient value; The review engine unit generates a target second quantization tag according to the quantization quotient value; The first quantization tag and the second quantization tag form the quantization tag.

10. A general artificial intelligence-based text content analysis and reasoning system according to claim 9, characterized in that, When the quantization quotient value of each task is less than 1, the number of second quantization tags is zero, and at this time, the second quantization tag is zero; When there is a quantization quotient value greater than 1 among the quantization quotient values of all tasks, the number of the second quantization tags is greater than zero: when the quantization quotient value of the task is an integer greater than 1, the review engine unit generates the second quantization tag with the number equal to the quantization quotient value of the task minus 1, that is, the second quantization tag of the task; when the quantization quotient value of the task is a non-integer greater than 1, the review engine unit generates the second quantization tag with the number equal to the quantization quotient value of the task, that is, the second quantization tag of the task; The second quantization tags of all tasks form the target second quantization tag.

Images