Literature work set analysis method and system

Abstract

The invention discloses a literature work set analysis method and system, relates to the technical field of AI natural language processing, and solves the technical problems that literature work understanding only stays on the surface, and literature works are not deeply researched and analyzed. Various redundant format information is automatically removed and converted into regular preprocessed texts, so that the data acquisition efficiency is improved, a foundation is built for subsequent analysis, errors caused by format problems are reduced, the accuracy of analysis results is ensured, high-frequency vocabularies are accurately locked by means of a scientific word segmentation tool and a rigorous word frequency statistical algorithm, and the accuracy of the high-frequency vocabularies is improved. And innovatively classifying and counting the proportion according to vocabulary semantics, so that the core theme of the works can be quickly positioned, the dynamic change of the theme in each chapter and paragraph can be clearly presented, readers and researchers are assisted to quickly grasp key contents of the works, and a large amount of manual sorting time is saved.

Description

Technical Field

[0001] The present invention relates to the field of AI natural language processing technology, and specifically to a method and system for analyzing a collection of literary works. Background Art

[0002] With the rapid development of the digital age, the number of literary works has exploded, including the vast electronic versions of classic masterpieces in electronic libraries, the massive amount of online articles updated daily on online literary platforms, and various original works published on writers' personal websites. These rich literary resources bring opportunities to readers, researchers and creators, but also trigger new challenges.

[0003] The patent with publication number CN115221861A discloses a method and system for analyzing a collection of literary works, which is implemented using NLP (Natural Language Processing) technology. The method includes: selecting a collection of literary works to be analyzed, pre-processing it, and then importing it into a programming platform; according to the analysis target, the user calls at least one NLP tool module in the NLP tool module set pre-deployed and displayed on the programming platform through instructions to analyze the collection of literary works to be analyzed.

[0004] However, some existing tools can only perform simple word frequency statistics, and cannot deeply analyze the semantic information and emotional tendencies behind the words; some tools can do basic grammar verification, but cannot combine vocabulary features, conduct a comprehensive analysis of grammatical elements such as sentence patterns and sentence length distribution from the perspective of literary creation, and explore the unique style and rhythm of the work. In addition, existing tools often view vocabulary and grammar in isolation, and rarely organically integrate the two to fully interpret the deep meaning, theme direction and emotional context contained in literary works. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method and system for analyzing a collection of literary works, which solves the problem of only superficial understanding of literary works and lack of in-depth research and analysis of literary works.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for analyzing a collection of literary works, the method specifically comprising the following steps: Step 1: Obtain the literary works to be analyzed and record them as the analysis object; Step 2: Clean the text format corresponding to the obtained analysis object to obtain a preprocessed analysis object; Step 3: Perform lexical feature analysis on the obtained preprocessed analysis object, extract the number of words, and obtain high-frequency words based on all the words. At the same time, analyze the distribution of high-frequency words to obtain the high-frequency word distribution results; Step 4: Analyze the grammatical features of the preprocessed analysis object by segmenting the vocabulary sentences and analyzing the preprocessed analysis object in combination with the length distribution of different vocabulary sentences, and generate a grammatical analysis result by combining the sentence patterns of the vocabulary sentences; Step 5: Conduct a comprehensive analysis of the literary works based on the high-frequency vocabulary distribution results and grammatical analysis results. The analysis is performed by analyzing the matching degree between the high-frequency vocabulary distribution and the vocabulary sentences, and a comprehensive judgment is made in combination with the grammatical structure to generate the analysis results.

[0007] As a further solution of the present invention, the specific method of performing vocabulary feature analysis on the prediction analysis object in step 3 is: Obtain a preprocessing analysis object, and use a word segmentation tool to perform word segmentation on the preprocessing analysis object, and record the obtained vocabulary number as i, and i=1, 2, ..., j, where j represents the number of vocabulary. At the same time, obtain the number of occurrences corresponding to vocabulary i, record it as Ci, and sort the vocabulary i according to the number of occurrences Ci, then calculate the appearance frequency corresponding to vocabulary i, record it as Pi, and compare the appearance frequency Pi with a preset value Py, and the specific value of the preset value Py is set by the operator.

[0008] As a further solution of the present invention, the specific method of comparing the occurrence frequency with the preset value in step 3 is: The vocabulary i whose appearance frequency Pi is greater than the preset value Py is defined as a high-frequency vocabulary, and the vocabulary i whose appearance frequency Pi is less than the preset value Py is defined as a normal vocabulary, then the preprocessing analysis object is divided into text segments, and the text segments are labeled as n, and n=1, 2, ..., m, where m represents the number of text segments, and the number of high-frequency words corresponding to the text segment n is obtained at the same time, and then the vocabulary type of the obtained high-frequency words is obtained, and classified according to the vocabulary type to obtain a classified high-frequency vocabulary group; Then, the classified high-frequency vocabulary group is obtained, and the corresponding number of high-frequency vocabulary of the same type is obtained at the same time, and the high-frequency vocabulary of the same type is sorted from large to small to generate the high-frequency vocabulary distribution result.

[0009] As a further solution of the present invention, the specific method of analyzing the grammatical features of the preprocessing analysis object in step 4 is: Obtain a preprocessing analysis object, and divide the preprocessing analysis object into text segments n, where n=1, 2, ..., m. Then divide the text segment n into sentences according to punctuation marks, and obtain any group of text segments as the target object for analysis, obtain the number of sentences in the target object and mark it as o, where o=1, 2, ..., p, and obtain the vocabulary length of sentence o, and obtain the number of sentences o corresponding to different vocabulary lengths. Then obtain the ending punctuation mark corresponding to sentence o, and classify the same type of symbol sentences according to the ending punctuation mark, and obtain the number of the same type of symbol sentences, and analyze according to the number of the same type of symbol sentences.

[0010] As a further solution of the present invention, the specific method of analyzing according to the number of symbolic statements of the same type in step 4 is: The number of symbolic sentences of the same type is obtained, and the corresponding number ratio of symbolic sentences of the same type is calculated, and they are sorted from large to small according to the number ratio. Similarly, all symbolic sentences of the same type are processed in the same way, and then the symbolic sentence of the same type with the largest number ratio is obtained. At the same time, the distribution of high-frequency words in the symbolic sentences of the same type is obtained, and the emotional words of the high-frequency words are matched according to the HowNet emotional dictionary.

[0011] As a further solution of the present invention, the specific method of matching the sentiment words of the high-frequency words in step 4 is: Compare the high-frequency words with the words in the HowNet sentiment dictionary one by one. If the high-frequency words are completely consistent with the sentiment words in the dictionary, it is considered a successful match. If the high-frequency words are inconsistent with the sentiment words in the dictionary, a synonym dictionary or word form transformation rules are used for matching; According to the sentiment polarity annotations of the matched sentiment words in the HowNet sentiment dictionary, the sentiment tendency of each word is determined, and specifically classified into three sentiment tendencies: positive, negative and neutral. At the same time, the corresponding quantities of the three sentiment tendencies are obtained respectively, and the overall sentiment tendency of the target object is determined based on the comparison of the quantities to generate the grammatical analysis results.

[0012] As a further solution of the present invention, the specific method of comprehensively analyzing the high-frequency vocabulary distribution results and the grammatical analysis results of the literary works in step 5 is: Obtain all text fragments, and calculate the matching degree between the high-frequency vocabulary distribution in the text fragments and the grammatical analysis results, obtain any group of text fragments, and simultaneously obtain the vocabulary sentences in the text fragments, and obtain the corresponding high-frequency vocabulary in the vocabulary sentences, and then calculate the vocabulary proportion value corresponding to the high-frequency vocabulary, and similarly calculate the vocabulary proportion values ​​of all high-frequency vocabulary, and sum the vocabulary proportion values ​​of the same high-frequency vocabulary, and at the same time identify the vocabulary semantics corresponding to the high-frequency vocabulary, and then classify the high-frequency vocabulary with the same vocabulary semantics into the same type, and calculate the sum of all vocabulary proportion values ​​with the same vocabulary semantics as the vocabulary judgment value; Then, the high-frequency words corresponding to the maximum vocabulary judgment value are obtained, and the text fragments are annotated based on the obtained high-frequency words, and the analysis results are generated at the same time.

[0013] A literary collection analysis system includes a data acquisition communication module, a preprocessing communication module, a feature extraction communication collaboration module, a comprehensive analysis communication module, and an analysis information output communication module; A data acquisition communication module, which is used to acquire the literary works to be analyzed and transmit the acquired literary works to the preprocessing communication module; A preprocessing communication module, which is used to preprocess the acquired literary works, obtain a preprocessing analysis object by cleaning the text format, and transmit the preprocessing analysis object to the feature extraction communication coordination module; A feature extraction communication coordination module, which is used to process the acquired pre-processing analysis object, extract the number of words, and obtain high-frequency words based on all word analyses, and analyze the distribution of high-frequency words to obtain high-frequency word distribution results, and segment the word sentences, and analyze the pre-processing analysis object in combination with the length distribution of different word sentences, and perform a comprehensive analysis in combination with the sentence patterns of the word sentences to generate a grammatical analysis result, and then transmit the high-frequency word distribution result and the grammatical analysis result to the comprehensive analysis communication module; A comprehensive analysis communication module, which is used to conduct a comprehensive analysis of the literary works based on the obtained high-frequency vocabulary distribution results and grammatical analysis results, analyze the matching degree between the high-frequency vocabulary distribution and the vocabulary sentences, and make a comprehensive judgment based on the grammatical structure to generate the analysis results, and transmit the analysis results to the analysis information output communication module; The analysis information output communication module is used to display the acquired analysis results to the corresponding operator.

[0014] The present invention provides a method and system for analyzing a collection of literary works. Compared with the prior art, it has the following beneficial effects: The present invention accurately captures literary works from multiple channels, automatically removes various redundant format information, and converts them into regular pre-processed texts. This not only improves data acquisition efficiency, but also lays a solid foundation for subsequent analysis, reduces errors caused by format problems, and ensures the accuracy of analysis results. With the help of scientific word segmentation tools and rigorous word frequency statistical algorithms, high-frequency words are accurately locked, and innovatively classified and statistically accounted for by word semantics, so that the core theme of the work can be quickly located, and the dynamic changes of the theme in each chapter and paragraph can be clearly presented, helping readers and researchers to quickly grasp the key content of the work and saving a lot of manual sorting time. Secondly, the length of sentences, sentence structure and punctuation are comprehensively considered to deeply analyze the grammatical features of the works. By analyzing the rhythm created by the combination of long and short sentences, and the emotional attitudes contained in different sentence patterns, the style of the works is accurately revealed; combined with the HowNet emotional dictionary to match high-frequency vocabulary emotions, the emotional tendency of the text is objectively determined, making the style and emotional interpretation of literary works more objective and professional. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a step method diagram of the present invention; Figure 2 This is a schematic diagram of the system module of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] For example, see Figure 1 , the present application provides a method for analyzing a collection of literary works, the method specifically comprising the following steps: Step 1: Obtain the literary works that need to be analyzed and record them as the analysis objects. The specific literary works can be collected from electronic libraries, online literary platforms, writers' personal websites and other channels, and the obtained literary works are converted into text format for storage.

[0018] Step 2: Clean the text format corresponding to the obtained analysis object to obtain a preprocessed analysis object.

[0019] The analysis object is obtained, and the analysis object here is an analysis object in text format. At the same time, the text data of the analysis object is formatted and cleaned. Specifically, extra spaces, line breaks, and special symbols (such as copyright symbols, page number marks, etc.) in the text are removed. For example, some e-books may have interference from the typesetting format, such as extra indents at the beginning of each paragraph or special separators between chapters. These need to be cleaned up to make the text format unified. After format cleaning, the preprocessed analysis object is obtained.

[0020] Step 3: Perform lexical feature analysis on the obtained preprocessed analysis object, extract the number of words, and obtain high-frequency words based on all the words. At the same time, analyze the distribution of high-frequency words to obtain the high-frequency word distribution results.

[0021] Get the preprocessing analysis object, and use the word segmentation tool (jieba word segmentation) to segment the preprocessing analysis object, and record the obtained vocabulary number as i, and i=1, 2, ..., j, where j represents the number of vocabulary, and obtain the number of occurrences corresponding to vocabulary i as Ci, and sort the vocabulary i according to the number of occurrences Ci, then calculate the occurrence frequency corresponding to vocabulary i as Pi, and compare the occurrence frequency Pi with the preset value Py, and the specific value of the preset value Py is set by the operator. For example, a vocabulary with an occurrence frequency higher than a certain percentage (such as 1%) or higher than a specific number of times (such as 10 times) is regarded as a high-frequency vocabulary. For example, if after word segmentation we get the three words "sky", "very", and "blue", then j=3, and the corresponding i numbers of these three words are 1 (corresponding to "sky"), 2 (corresponding to "very"), and 3 (corresponding to "blue"), respectively. Then, for each word labeled i, we need to count the number of times it appears in the entire text, and record this number as Ci. For example, in a short article simply describing the weather, "sky" appears 5 times, "very" appears 3 times, and "blue" appears 4 times. Then for the word "sky", i=1, C1=5; for the word "very", i=2, C2=3; for the word "blue", i=3, C3=4; Next, calculate the frequency of each word i in the entire text, and record it as Pi. Assuming that the short text has a total of 20 words (i.e. N=j=20), then for the word "sky", its frequency of occurrence is P1=0.25; for word 5 "very", P0.15; for the word "blue", P3=0.2; The vocabulary i with an appearance frequency Pi greater than a preset value Py is defined as a high-frequency vocabulary, and the vocabulary i with an appearance frequency Pi less than a preset value Py is defined as a normal vocabulary, and then the preprocessing analysis object is divided into text segments, and the division method here includes dividing according to themes or chapters, and the text segments are labeled as n, and n=1, 2, ..., m, where m represents the number of text segments, and at the same time, the number of high-frequency words corresponding to the text segment n is obtained, and then the vocabulary type of the obtained high-frequency words is obtained, and classified according to the vocabulary type to obtain a classified high-frequency vocabulary group; Then, the classified high-frequency vocabulary group is obtained, and the corresponding number of high-frequency vocabulary of the same type is obtained at the same time, and the high-frequency vocabulary of the same type is sorted from large to small to generate the high-frequency vocabulary distribution result.

[0022] Step 4: Analyze the grammatical features of the preprocessed analysis object by segmenting the vocabulary sentences and analyzing the preprocessed analysis object in combination with the length distribution of different vocabulary sentences. At the same time, perform a comprehensive analysis in combination with the sentence patterns of the vocabulary sentences to generate a grammatical analysis result.

[0023] Obtain a preprocessing analysis object, and divide the preprocessing analysis object into text segments denoted as n, where n=1, 2, ..., m, and the text segments here are divided in the same way as in step three, and the specific text segments obtained are the same as the text segments in step three, then divide the text segment n into sentences according to punctuation marks, and the punctuation marks here are such as periods, question marks, exclamation marks, etc., and obtain any group of text segments as target objects for analysis, obtain the number of sentences in the target object and mark it as o, where o=1, 2, ..., p, and obtain the vocabulary length of sentence o, and the vocabulary length here is specifically represented by the corresponding vocabulary number, for example, in an article, the length is 0-5 words There are 20 sentences with 10 words, 30 sentences with 6-10 words, and so on, and the number of sentences o corresponding to different word lengths is obtained. Long sentences can express complex thoughts and delicate emotions, and slow down the rhythm of the article. For example, in stream-of-consciousness novels, longer sentences often appear to reflect the continuity and complexity of the character's thinking. Short sentences speed up the rhythm and give people a concise and brisk feeling. They are often used in tense plots or to emphasize key content. For example, in the climax of a suspense novel, a large number of short sentences may be used to enhance the tension. Then, the ending punctuation mark corresponding to sentence o is obtained, and the same type of symbol sentences are classified according to the ending punctuation mark, and the number of the same type of symbol sentences is obtained; The number of symbolic sentences of the same type is obtained, and the corresponding number ratio of symbolic sentences of the same type is calculated, and they are sorted from large to small according to the number ratio. Similarly, all symbolic sentences of the same type are processed in the same way, and then the symbolic sentence of the same type corresponding to the largest number ratio is obtained. At the same time, the distribution of high-frequency words in the symbolic sentences of the same type is obtained, and the emotional words of the high-frequency words are matched according to the HowNet emotional dictionary. The specific matching process is as follows: Compare the high-frequency words with the words in the HowNet sentiment dictionary one by one. If the high-frequency words are completely consistent with the sentiment words in the dictionary, it is considered a successful match. If the high-frequency words are inconsistent with the sentiment words in the dictionary, use the synonym dictionary or word form transformation rules to match them. For example, some words may not be exactly the same as the sentiment words in the dictionary in terms of form or expression, but have similar meanings; According to the sentiment polarity annotation of the matched sentiment words in the HowNet sentiment dictionary, the sentiment tendency of each word is determined, which is generally divided into three types: positive, negative and neutral. At the same time, the number corresponding to the three sentiment tendencies is obtained respectively, and the number here is the number of high-frequency words. The overall sentiment tendency of the target object is determined by comparison based on the number, and the grammatical analysis results are generated; Take a short article describing campus life as an example. After preliminary sorting, the high-frequency words are "laughter", "exam", "tiredness", "friendship" and "disappointment". Comparing with the emotion dictionary of CNKI, "laughter" and "friendship" are classified as positive emotion words, "tiredness" and "disappointment" are included in the negative emotion category, and "exam" is regarded as a neutral word. Statistics show that the number of positive emotion high-frequency words is 2, negative is 2, and neutral is 1. If the number of positive emotion words is greater than the number of negative emotion words, then the overall emotional tendency of the text may be positive; otherwise, it may be negative.

[0024] All text fragments are processed in this way to obtain corresponding grammatical analysis results.

[0025] Step 5: Conduct a comprehensive analysis of the literary works based on the high-frequency vocabulary distribution results and grammatical analysis results. The analysis is performed by analyzing the matching degree between the high-frequency vocabulary distribution and the vocabulary sentences, and a comprehensive judgment is made in combination with the grammatical structure to generate the analysis results.

[0026] Get all the text fragments and calculate the matching degree between the high-frequency vocabulary distribution in the text fragments and the grammatical analysis results. The specific calculation process is as follows: Obtain any group of text fragments, obtain vocabulary sentences in the text fragments, and obtain the corresponding high-frequency words in the vocabulary sentences, then calculate the vocabulary proportion value corresponding to the high-frequency words, and the vocabulary proportion value here is the proportion of the number of all high-frequency words and vocabulary sentences, and similarly calculate the vocabulary proportion values ​​of all high-frequency words, and sum the vocabulary proportion values ​​of the same high-frequency words, and identify the vocabulary semantics corresponding to the high-frequency words, then classify the high-frequency words with the same vocabulary semantics into the same type, and calculate the sum of all vocabulary proportion values ​​with the same vocabulary semantics as the vocabulary judgment value; Then, the high-frequency words corresponding to the maximum vocabulary judgment value are obtained, and the text fragments are annotated based on the obtained high-frequency words, and the analysis results are generated at the same time.

[0027] For example 2, please refer to Figure 2 The present application provides a literary collection analysis system, including: a data acquisition communication module, a preprocessing communication module, a feature extraction communication collaboration module, a comprehensive analysis communication module and an analysis information output communication module, and in combination with the attached Figure 2 It can be seen that all the above functional communication modules are electrically connected in a unidirectional manner.

[0028] A data acquisition communication module, which is used to acquire the literary works to be analyzed and transmit the acquired literary works to the preprocessing communication module; A preprocessing communication module, which is used to preprocess the acquired literary works, obtain a preprocessing analysis object by cleaning the text format, and transmit the preprocessing analysis object to the feature extraction communication coordination module, and the processing method here is the same as the processing process of step 2 in embodiment 1; A feature extraction communication coordination module, which is used to process the acquired pre-processing analysis object, extract the number of words, and obtain high-frequency words based on all word analyses, and analyze the distribution of high-frequency words to obtain high-frequency word distribution results, and the processing method here is the same as the processing process of step three in embodiment one, and the vocabulary sentences are segmented, and the pre-processing analysis object is analyzed in combination with the length distribution of different vocabulary sentences, and the sentence pattern of the vocabulary sentences is comprehensively analyzed to generate grammatical analysis results, and the processing method here is the same as the processing process of step four in embodiment one, and then the high-frequency word distribution results and grammatical analysis results are transmitted to the comprehensive analysis communication module; A comprehensive analysis communication module, which is used to conduct a comprehensive analysis of the literary works based on the obtained high-frequency vocabulary distribution results and grammatical analysis results, analyze the matching degree between the high-frequency vocabulary distribution and the vocabulary sentences, and make a comprehensive judgment based on the grammatical structure to generate the analysis results, and transmit the analysis results to the analysis information output communication module; The analysis information output communication module is used to display the acquired analysis results to the corresponding operator.

[0029] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0030] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for analyzing a collection of literary works, characterized in that: The method specifically comprises the following steps: Step 1: Obtain the literary works to be analyzed and record them as the analysis object; Step 2: Clean the text format corresponding to the obtained analysis object to obtain a preprocessed analysis object; Step 3: Perform lexical feature analysis on the obtained preprocessed analysis object, extract the number of words, and obtain high-frequency words based on all the words. At the same time, analyze the distribution of high-frequency words to obtain the high-frequency word distribution results; Step 4: Analyze the grammatical features of the preprocessed analysis object by segmenting the vocabulary sentences and analyzing the preprocessed analysis object in combination with the length distribution of different vocabulary sentences, and generate a grammatical analysis result by combining the sentence patterns of the vocabulary sentences; Step 5: Conduct a comprehensive analysis of the literary works based on the high-frequency vocabulary distribution results and grammatical analysis results. The analysis is performed by analyzing the matching degree between the high-frequency vocabulary distribution and the vocabulary sentences, and a comprehensive judgment is made in combination with the grammatical structure to generate the analysis results.

2. A method for analyzing a collection of literary works according to claim 1, characterized in that: The specific method of performing lexical feature analysis on the prediction analysis object in step 3 is: Obtain a preprocessing analysis object, and use a word segmentation tool to perform word segmentation on the preprocessing analysis object, and record the obtained vocabulary number as i, and i=1, 2, ..., j, where j represents the number of vocabulary. At the same time, obtain the number of occurrences corresponding to vocabulary i, record it as Ci, and sort the vocabulary i according to the number of occurrences Ci, then calculate the appearance frequency corresponding to vocabulary i, record it as Pi, and compare the appearance frequency Pi with a preset value Py, and the specific value of the preset value Py is set by the operator.

3. A method for analyzing a collection of literary works according to claim 2, characterized in that: The specific method of comparing the occurrence frequency with the preset value in step 3 is: The vocabulary i whose appearance frequency Pi is greater than the preset value Py is defined as a high-frequency vocabulary, and the vocabulary i whose appearance frequency Pi is less than the preset value Py is defined as a normal vocabulary, then the preprocessing analysis object is divided into text segments, and the text segments are labeled as n, and n=1, 2, ..., m, where m represents the number of text segments, and the number of high-frequency words corresponding to the text segment n is obtained at the same time, and then the vocabulary type of the obtained high-frequency words is obtained, and classified according to the vocabulary type to obtain a classified high-frequency vocabulary group; Then, the classified high-frequency vocabulary group is obtained, and the corresponding number of high-frequency vocabulary of the same type is obtained, and the high-frequency vocabulary of the same type is sorted from large to small to generate the high-frequency vocabulary distribution result.

4. A method for analyzing a collection of literary works according to claim 1, characterized in that: The specific method of analyzing the grammatical features of the preprocessing analysis object in step 4 is: Obtain a preprocessing analysis object, and divide the preprocessing analysis object into text segments n, where n=1, 2, ..., m. Then divide the text segment n into sentences according to punctuation marks, and obtain any group of text segments as the target object for analysis, obtain the number of sentences in the target object and mark it as o, where o=1, 2, ..., p, and obtain the vocabulary length of sentence o, and obtain the number of sentences o corresponding to different vocabulary lengths. Then obtain the ending punctuation mark corresponding to sentence o, and classify the same type of symbol sentences according to the ending punctuation mark, and obtain the number of the same type of symbol sentences, and analyze according to the number of the same type of symbol sentences.

5. A method for analyzing a collection of literary works according to claim 4, characterized in that: The specific method of analyzing the number of symbolic statements of the same type in step 4 is: The number of symbolic sentences of the same type is obtained, and the corresponding number ratio of symbolic sentences of the same type is calculated, and they are sorted from large to small according to the number ratio. Similarly, all symbolic sentences of the same type are processed in the same way, and then the symbolic sentence of the same type with the largest number ratio is obtained. At the same time, the distribution of high-frequency words in the symbolic sentences of the same type is obtained, and the emotional words of the high-frequency words are matched according to the HowNet emotional dictionary.

6. A method for analyzing a collection of literary works according to claim 5, characterized in that: The specific method of matching the sentiment words of high-frequency words in step 4 is: Compare the high-frequency words with the words in the HowNet sentiment dictionary one by one. If the high-frequency words are completely consistent with the sentiment words in the dictionary, it is considered a successful match. If the high-frequency words are inconsistent with the sentiment words in the dictionary, a synonym dictionary or word form transformation rules are used for matching; According to the sentiment polarity annotations of the matched sentiment words in the HowNet sentiment dictionary, the sentiment tendency of each word is determined, and specifically classified into three sentiment tendencies: positive, negative and neutral. At the same time, the corresponding quantities of the three sentiment tendencies are obtained respectively, and the overall sentiment tendency of the target object is determined based on the comparison of the quantities to generate the grammatical analysis results.

7. A method for analyzing a collection of literary works according to claim 1, characterized in that: The specific method of comprehensively analyzing the high-frequency vocabulary distribution results and grammatical analysis results of the literary works in step 5 is: Obtain all text fragments, and calculate the matching degree between the high-frequency vocabulary distribution in the text fragments and the grammatical analysis results, obtain any group of text fragments, and simultaneously obtain the vocabulary sentences in the text fragments, and obtain the corresponding high-frequency vocabulary in the vocabulary sentences, and then calculate the vocabulary proportion value corresponding to the high-frequency vocabulary, and similarly calculate the vocabulary proportion values ​​of all high-frequency vocabulary, and sum the vocabulary proportion values ​​of the same high-frequency vocabulary, and at the same time identify the vocabulary semantics corresponding to the high-frequency vocabulary, and then classify the high-frequency vocabulary with the same vocabulary semantics into the same type, and calculate the sum of all vocabulary proportion values ​​with the same vocabulary semantics as the vocabulary judgment value; Then, the high-frequency words corresponding to the maximum vocabulary judgment value are obtained, and the text fragments are annotated based on the obtained high-frequency words, and the analysis results are generated at the same time.

8. A literary collection analysis system, used to execute a literary collection analysis method according to any one of claims 1 to 7, characterized in that: It includes a data acquisition communication module, a preprocessing communication module, a feature extraction communication coordination module, a comprehensive analysis communication module and an analysis information output communication module; A data acquisition communication module, which is used to acquire the literary works to be analyzed and transmit the acquired literary works to the preprocessing communication module; A preprocessing communication module, which is used to preprocess the acquired literary works, obtain a preprocessing analysis object by cleaning the text format, and transmit the preprocessing analysis object to the feature extraction communication coordination module; A feature extraction communication coordination module, which is used to process the acquired pre-processing analysis object, extract the number of words, and obtain high-frequency words based on all word analyses, and analyze the distribution of high-frequency words to obtain high-frequency word distribution results, and segment the word sentences, and analyze the pre-processing analysis object in combination with the length distribution of different word sentences, and perform a comprehensive analysis in combination with the sentence patterns of the word sentences to generate a grammatical analysis result, and then transmit the high-frequency word distribution result and the grammatical analysis result to the comprehensive analysis communication module; A comprehensive analysis communication module, which is used to conduct a comprehensive analysis of the literary works based on the obtained high-frequency vocabulary distribution results and grammatical analysis results, analyze the matching degree between the high-frequency vocabulary distribution and the vocabulary sentences, and make a comprehensive judgment based on the grammatical structure to generate the analysis results, and transmit the analysis results to the analysis information output communication module; The analysis information output communication module is used to display the acquired analysis results to the corresponding operator.

Images