Network big data text analysis multi-dimensional clustering theme evolution model

Abstract

According to the network big data text analysis multi-dimensional clustering topic evolution model, different topic dimensions are selected and text analysis and text classification are carried out through the doubt degree and the log-likelihood function evaluation index, and on one hand, the optimal topic dimension of the topic evolution model is estimated; a method for judging theme dimensions in a corpus based on doubt degrees and log-likelihood dynamic functions is adopted, changes between the doubt degrees and the log-likelihood dynamic functions under different theme dimensions are calculated, and different optimal theme dimensions are obtained; 2, evaluating modeling effects under different theme dimensions by utilizing a visualization method from the perspective of text analysis on the basis of a theme evolution model of text analysis; and thirdly, based on a topic evolution model of text classification, text representation is constructed through topic distribution on the basis of a distribution relationship between documents and topics, and the text representation is used as a feature vector to determine an optimal topic dimension, so that the precision and efficiency of network big data text analysis and classification are high.

Description

Technical Field

[0001] The present application relates to a big data text analysis and classification model, and in particular to a multi-dimensional clustering topic evolution model for network big data text analysis, and belongs to the technical field of data text classification. Background Art

[0002] Emerging smart cities and intelligent cities are developing rapidly, regulating and managing cities in an information-based way to provide residents with more comprehensive and appropriate services. A mobile terminal can easily access a variety of information, such as map service information, news media information, advertising promotions, etc. With the development and popularization of the Internet, the surge in information has forced information managers to face the management difficulties brought about by big data. In the face of the arrival of big data, effective organization and management have become the current focus and difficulty.

[0003] Although various complex and varied information carriers have appeared in people's current situation, the most common information is still expressed in text. Not only is text the most direct way for people to accept information, but text expression is also the least consuming. How to effectively manage text information and how to intuitively present the information expressed by text are becoming more and more hot issues in natural language processing.

[0004] When processing text, the first thing to do is to express the text in a vectorized way, and convert the text information into an object that can be processed by the computer. The more mature existing technology uses the vector space model to represent the text structure, converts the study of text into the study of word vectors, and then builds a classifier through classification algorithms such as the Bayesian classification algorithm and the support vector machine algorithm. By learning knowledge on a set of text samples of known categories, unknown texts can be judged. Of course, these methods still need to be improved. First, the parts of speech that make up the document are different, and the number of words is related to the length of the document. A long document contains many words, resulting in a huge dimension of words in the final model; secondly, the basis of the classification problem is the semantics of the text, and the vector space model represents the distribution of words. The difference between the two may cause inaccurate classification results through the classifier. On the one hand, through the dimensionality reduction technology, the dimension of the vector space model representing the text is compressed to a low-dimensional space; on the other hand, when extracting text words, the words that can better express the text content are retained to improve the accuracy of text classification. Through topic modeling, the distribution of "topics" in documents can be learned, and each "topic" is the distribution of words. In the topic space, a topic is composed of a group of words, so the number of topics is much smaller than words. Representing texts through the topic space can not only solve the problem of high dimensionality of texts represented by word vectors, but also improve the problem that scattered words cannot express the semantics of texts. However, how to more intuitively understand the meaning of "topics" in the topic evolution model and how to understand the content of the document set in a macro sense have become the primary problems to be solved by topic models such as data evolution.

[0005] The problems that the existing network text parsing classification model needs to solve and the key technical difficulties of this application include: (1) The existing technology cannot effectively manage text information and intuitively present the information expressed by the text. First, the parts of speech that make up the document are different, and the number of words is related to the length of the document. A long document contains many words, resulting in a huge dimension of the word representation in the final model; second, the basis of the classification problem is the semantics of the text, while the vector space model represents the distribution of words. The difference between the two may cause inaccuracy in the classification results of the classifier. The existing technology lacks targeted dimensionality reduction technology and cannot compress the dimension of the vector space model representing the text into a low-dimensional space; on the other hand, when extracting text vocabulary, it is impossible to retain the vocabulary that expresses the text content to improve the accuracy of text classification. Through topic modeling, the distribution of "topics" in documents can be learned, and each "topic" is the distribution of words. However, the existing technology cannot intuitively understand the meaning of "topics" in the topic evolution model, and cannot understand the content of the document set from a macro perspective, resulting in poor accuracy and low efficiency in network big data text parsing and classification.

[0006] (2) The topic model describes the documents in the corpus as a distribution of topics, and topics are distributions of terms. When faced with text parsing problems, there is often a certain degree of ambiguity about which terms can best express the text. When faced with text classes, the spatial dimensions of describing the text are generally very high. The more topic dimensions there are, the more difficult and computationally complex the classification will be. If the topic dimensions are too few, a lot of important information will be discarded. Existing technologies cannot make a good trade-off. There is a lack of optimal topic dimension estimation for the topic evolution model, and it is impossible to combine the language model evaluation indicators with the structural characteristics of the data evolution model itself. There is a lack of a method to determine the topic dimension in the corpus based on doubt and log-likelihood dynamic functions. It is impossible to calculate the changes between doubt and log-likelihood dynamic functions under different topic dimensions, and it is impossible to obtain different optimal topic dimensions.

[0007] (3) The existing technology lacks a topic evolution model based on text parsing, and is unable to use visualization methods to evaluate the modeling effects under different topic dimensions from the perspective of text parsing. It lacks an analysis method that highlights high-frequency words in the text parsing of "document-term", lacks a text parsing method based on data evolution and a term analysis method in the topic space, and is unable to simultaneously determine the optimal topic dimension through the distribution of "topic-topic" and "topic-term" after modeling; the existing technology lacks a topic evolution model based on text classification, lacks a text representation based on the distribution relationship between documents and topics through topic distribution, and is unable to determine the optimal topic dimension by comparing different topic modeling through text classification. It adopts a single selection method based on doubt degree, and does not optimize the topic evolution model through the semantic expression of high-probability terms and specific comparison results from the perspective of text parsing and text classification, and is unable to evaluate the modeling effect of selecting topic dimensions based on different indicators. Summary of the invention

[0008] This application selects different topic dimensions and performs text parsing and text classification through the evaluation indicators of doubt and log-likelihood function. First, the optimal topic dimension of the topic evolution model is estimated. Combining the evaluation indicators of the language model with the structural characteristics of the data evolution model itself, a method based on doubt and log-likelihood dynamic function is adopted to judge the topic dimension in the corpus, and the changes between doubt and log-likelihood dynamic function under different topic dimensions are calculated. The topic dimension with the lowest doubt and the topic dimension with the largest log-likelihood dynamic function are obtained through M-layer cross-validation to obtain different optimal topic dimensions; second, a topic evolution model based on text parsing is used to evaluate the modeling effect under different topic dimensions from the perspective of text parsing using visualization methods; third, a topic evolution model based on text classification is based on the distribution relationship between documents and topics, and a text representation is constructed through the topic distribution, and this is used as a feature vector. Different topic models are modeled through text classification comparison to determine the optimal topic dimension. The accuracy and efficiency of network big data text parsing and classification are high.

[0009] In order to achieve the above technical effects, the technical solutions adopted in this application are as follows: The multi-dimensional clustering topic evolution model for network big data text analysis builds an evaluation method for estimating topic dimensions in data evolution models from the perspective of text analysis and text classification. By calculating the doubt degree and log-likelihood dynamic function under different topic dimensions, the topic dimension is estimated in a data-driven way. Then, based on the results of corpus modeling using the estimated topic dimension, the optimal topic dimension is judged in combination with text analysis and text classification. Through visualization under the topic evolution model, the text content is intuitively displayed in the form of word clouds and bubble charts from the perspectives of word frequency and document, word frequency and classification, and topic distribution based on data evolution, a dynamic comparison of the modeling effects based on log-likelihood dynamic function and the modeling effects based on doubt degree is obtained. M1-Estimation of the optimal topic dimension of the topic evolution model: Combining the evaluation indicators of the language model with the structural characteristics of the data evolution model itself, a method based on doubt and log-likelihood dynamic function is used to judge the topic dimension in the corpus, and the changes between doubt and log-likelihood dynamic function under different topic dimensions are calculated. Through M-layer cross-validation, the topic dimension with the lowest doubt and the topic dimension with the largest log-likelihood dynamic function are obtained, thereby obtaining different optimal topic dimensions; M2-Topic evolution model based on text analysis: From the perspective of text analysis, the visualization method is used to evaluate the modeling effect under different topic dimensions. By highlighting the analysis method of high-frequency words in the comparison of "document-term" text analysis, a text analysis method based on data evolution and a term analysis method in the topic space are constructed. At the same time, the optimal topic dimension is determined by the distribution of "topic-topic" and "topic-term" after modeling; M3-Topic evolution model based on text classification: Based on the distribution relationship between documents and topics, text representation is constructed through topic distribution and used as feature vector. Different topics are modeled through text classification comparison to determine the optimal topic dimension. The topic evolution model is optimized through the semantic expression of high-probability terms and specific comparison results, and the modeling effect of the topic dimension is evaluated based on different indicators.

[0010] Preferably, the optimal topic dimension estimation of the data evolution model is: by setting different topics M, observing the current number of topics M, using multi-layer cross-validation doubt and log-likelihood dynamic functions, and determining the M value by comparing the visualization graph, which is the optimal topic dimension estimation value of the current corpus.

[0011] Preferably, the data evolution model: assuming that the test text set is represented by M, the term in the bag-of-words model is represented by w, and the logarithmic likelihood value describes the distribution of the term in the topic space, firstly, the prior parameters of the "document-term" matrix are given, and the prior distribution parameter values ​​are determined by using Gibbs sampling, and the corpus is divided into a training set and a test set, and the model parameters are trained by using the training set, and then the logarithmic likelihood dynamic function of the test set is calculated by using VEM estimation; Estimating the topic dimension is a process of continuous model comparison. It analyzes the changes in the mean of the model's doubt and the mean of the log-likelihood dynamic function when the topic dimension changes. The cross-validation method is used to test the data. When modeling the corpus, the text data is randomly divided into M groups of equal size. One group is selected as the test set. The remaining M-1 groups of data are modeled to train the topic evolution model. The trained topic evolution model is then used to model the test set. A set of doubt and log-likelihood dynamic functions are calculated. Repeating this process multiple times under the same conditions will obtain multiple sets of doubt and log-likelihood dynamic functions, and the mean of multiple sets of doubt and log-likelihood dynamic functions is output as the result to complete a multi-layer cross-validation. When evaluating the modeling effect of the topic evolution model, it is determined that the lower the doubt, the better, and the larger the log-likelihood dynamic function, the better. After a topic modeling, the doubt and log-likelihood dynamic function are calculated at the same time. A multi-layer cross-check is performed under the current model parameters to obtain the average doubt and log-likelihood dynamic function. Then, the doubt and log-likelihood dynamic function trends output by the topic dimension analysis are changed. The process of solving the average doubt and log-likelihood dynamic function is as follows: Step 1: Given the subject dimension; Step 2: Select the Gibbs topic modeling algorithm; Step 3: Solve the current modeling doubt and log-likelihood dynamic function; Step 4: Repeat the M-layer cross-validation 6 times and record the average doubt and log-likelihood dynamic function; The topic dimension of the topic evolution model is simulated and determined through the M-layer cross-validation method, and the topic value is determined by combining the two evaluation indicators of doubt degree and log-likelihood function value.

[0012] Preferably, a topic evolution model based on text parsing is used: by analyzing the text after topic modeling and evaluating the effect of modeling to obtain the optimal modeling topic dimension, a topic evolution model is first constructed based on the estimated topic dimension corpus modeling to express two sets of "document-topic" and "topic-term" relationships. One is to express the core of the corpus by extracting terms under the topic as keywords, and to give an evaluation through comparison based on the analysis of the text word frequency word cloud; the other is to use the topic similarity evaluation method between topic structures as an evaluation of topic modeling, and to evaluate the modeling effects of different topic dimensions from an intuitive and objective perspective from the data evolution vis visualization analysis.

[0013] Preferably, word cloud analysis based on text word frequency: in word cloud analysis, the size of word weight is measured based on text word frequency, and high-weight words are output as the subject words of the word cloud. The process of generating the word cloud includes two steps: first, the weight of each word is calculated for the preprocessed text or corpus; second, the size of the word in the composition space is estimated according to the size of the word weight, and the word cloud is output. When outputting the word cloud, the priority of words with different weights in the composition space is considered, and the stability and rationality of the word layout are met by using a compact layout method.

[0014] Preferably, text parsing based on word cloud: the corpus is modeled through the data evolution model, and the distribution of "topic-term" in the corpus is learned. According to the estimated numbers of two topics, the corpus is subjected to topic modeling with a topic dimension of 20 and a topic modeling with a topic dimension of 60, respectively, to form two groups of comparisons. In the modeled corpus, the distribution of terms under each topic is analyzed, and the terms under each topic are distributed according to the probability of p(w|z). Words with large probabilities are prioritized in each topic space, and it is approximately determined that words with high word frequencies in different topic spaces are prioritized in front. In the modeling of the two groups, 120 words with higher probabilities in the topic spaces are extracted respectively, and the word frequencies are counted, and then displayed in the form of word clouds.

[0015] Preferably, the visual analysis based on the topic evolution model: Through data evolution vis network visualization as a text parsing tool after topic modeling of different topic dimensions, through multi-scale analysis, the relative distance between objects is grasped, and the expression form of data in low-dimensional space is obtained. The distribution of corpus documents in the topic space and the distribution of terms in the topic space after data evolution modeling are analyzed, and the probability size information of terms is expressed by the ranking of terms. After preprocessing the text in the corpus, the text content of text_content is formed, and then the text is subject modeled on this basis. The modeling results are visualized through the topic evolution model, and the output results are formulated in the form of web pages by writing scripts; The ranking of terms is changed by the adjustable parameter a to help explain the topic. If a is close to 1, the terms at the top are those with higher frequency in the topic space; if a is closer to 0, the terms at the top are those with more unique characteristics in the topic space. By adjusting the size of a, the weight of the terms under the topic is changed, affecting the order of the terms. The value of a is determined by the following formula: Formula 1 Based on the relevance of a certain word topic, it is adjusted by a. By comparing the frequency of a given word in the corpus with the topic-specific frequency, it is quickly determined whether it is a highly relevant selected topic; θ represents the vector of topic distribution, p(θ) is the distribution of θ, N represents the number of words in the corpus, w_n is the nth word w, z_n represents the selected topic, p(z|θ) represents the probability of topic z under the topic distribution θ, p(w|z) represents the probability of word w under topic z, expressed as an M×V matrix, M represents the topic, V represents the word in the corpus, α is the parameter describing the distribution p(θ), affecting the distribution of the topic θ vector, β is the parameter describing the distribution p(w|z), affecting the distribution of the word under topic z; The relationship between topics is analyzed by multidimensional scaling, which projects the high-dimensional topic distribution into two-dimensional space, and describes the differences between topics by the size and distance of the bubble chart; From the perspective of the distribution of terms under a topic, the fewer topic structures a term has associated with it in the topic space, the more evenly the distribution of "topic-term" is. The modeling effect is evaluated by the relationship between different modeling topic dimensions and the topic similarity obtained after modeling. That is, when the average similarity between topics is the smallest, the model is optimal. The probability distribution of topics in the term space is used to represent the topic vector, the cosine distance is used to represent the difference of topics, and cosine(z i ,z j ) is smaller, the greater the difference in the subject space: Formula 2 The quality of the topic structure is characterized by the average similarity of the topic vectors: Formula 3 The modeling effects of the two are compared directly by comparing the average topic similarities under different topic dimension modeling, and the average similarities of all topics in the topic evolution model under different topic dimensions are compared. It is determined that when the topic dimension is 20, the difference of topic vectors in the topic space is greater, and it is determined that the modeling performance with a topic dimension of 20 is relatively better.

[0016] Preferably, a topic evolution model based on multi-dimensional clustering of text is used: the "document-topic" matrix obtained after data evolution modeling of the test set in the corpus is used as the training set of the nearest neighbor classifier, the test set is obtained by the same method and predictions are made through the nearest neighbor classifier, firstly, the text representation of the data evolution modeling is constructed by estimating different topic dimensions, and the pros and cons of the different topic dimensions are estimated through feedback from the classification effect; then, the optimal text representation of the data evolution modeling is compared with the text representation constructed by the vector space model.

[0017] Preferably, a text classification model based on data evolution dimensionality reduction: a method for classifying text by using a text representation after modeling a corpus using a data evolution model is constructed, and the specific steps are as follows: Step 1: Use the data evolution model to model the training document set and obtain the distribution of documents on topics as the vector representation of the documents; Step 2: Use the nearest neighbor algorithm to train the vector in step 1; Step 3: Use the results in step 1 to model the test document set and obtain the topic distribution in the test document set; Step 4: Represent the result as a text vector with a fixed number of text features; Step 5: Predict the category of the text vector using the classifier obtained in step 2.

[0018] Preferably, a text classification model based on the nearest neighbor is used: by representing the text in the corpus as a feature vector, the category of the unknown text is determined to be the same as the majority of the categories of the T adjacent samples in the feature item space in the classification. The number of nearest neighbor samples T and the distance function are known, the T value is the square root of the number of training sets, and the distance function is the Euclidean distance. The algorithm flow is as follows: Step 1: First, divide the corpus into a test document set and a training document set, and perform preprocessing; Step 2: Extract specific parts of speech for all document sets in the corpus and convert them into text representation based on word vectors; Step 3: Select consistent feature vectors based on text word frequency; Step 4: For the vector to be classified in the training document set, traverse the distance between the vector and the known classification vector; Step 5: Select the nearest vectors in T vector spaces and calculate the category weights of the majority vectors; Step 6: Label the unknown vector as the class with the largest weight for the majority class.

[0019] Compared with the prior art, the innovations and advantages of this application are: (1) This application selects different topic dimensions and performs text parsing and text classification through the evaluation indicators of doubt and log-likelihood function. First, it constructs the optimal topic dimension estimation of the topic evolution model, combines the evaluation indicators of the language model with the structural characteristics of the data evolution model itself, and adopts a method based on doubt and log-likelihood dynamic function to judge the topic dimension in the corpus, calculates the changes between doubt and log-likelihood dynamic function under different topic dimensions, and uses M-layer cross-validation to find the topic dimension with the lowest doubt and the topic dimension with the largest log-likelihood dynamic function, thereby obtaining different optimal topic dimensions; second, it constructs a topic evolution model based on text parsing. From the perspective of text parsing, The visualization method is used to evaluate the effect of modeling under different topic dimensions. By highlighting the analysis method of high-frequency words in the text analysis of "document-term", a text analysis method based on data evolution and a term analysis method in the topic space are constructed. At the same time, the optimal topic dimension is determined by the distribution of "topic-topic" and "topic-term" after modeling; thirdly, a topic evolution model based on text classification is constructed. Based on the distribution relationship between documents and topics, text representation is constructed through topic distribution, and this is used as a feature vector. Different topic modeling is carried out through text classification comparison to determine the optimal topic dimension. The accuracy and efficiency of network big data text analysis and classification are high.

[0020] (2) One of the innovative points of this application is that in the use of topic evolution model for text representation, it avoids the single selection method of using doubt degree, and proposes to use two judgment indicators, doubt degree and log-likelihood dynamic function, for selection. The cross-validation method is used to record the doubt degree and log-likelihood dynamic function of the model under different topic dimensions in a data-driven manner. The average change results of doubt degree and log-likelihood dynamic function are used as the basis for selecting topic dimensions. From the perspective of text analysis and text classification, the topic evolution model is optimized through the semantic expression of high-probability terms and specific comparison results, and the modeling effect of selecting topic dimensions based on different indicators is evaluated, which is conducive to obtaining the optimal topic dimension under different situations.

[0021] (3) This application proposes to evaluate the modeling effect of the data evolution model from the perspective of text parsing and text classification. In terms of text parsing, the modeling effect of the topic evolution model under different topic dimensions is explored, and the modeling effect is measured by word cloud analysis, potential topic distribution, and topic similarity, so as to evaluate the model of different topic dimension estimation in the topic evolution model. In terms of text classification, it is proposed to evaluate the modeling effect of the data evolution model from the perspective of text classification results. Existing text classification research focuses on the performance comparison between different classification methods. This application proposes to evaluate the modeling effect of the topic evolution model of different topic dimensions from the perspective of task results. In view of the estimation of multiple topic dimensions, the corpus is modeled and represented in different topic dimensions. The modeling effect under different topic dimensions is evaluated through the nearest classification results. Text classification based on topic dimensionality reduction has good feasibility and effectiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is the overall block diagram of the optimal topic dimension estimation for the data evolution model.

[0023] Figure 2 This is the overall framework diagram of the topic evolution model based on text parsing.

[0024] Figure 3 This is the overall framework diagram of the topic evolution model based on multi-dimensional text clustering.

[0025] Figure 4 It is a flowchart of text classification based on data evolution dimensionality reduction.

[0026] Figure 5 It is a flowchart of text classification based on the nearest neighbor. DETAILED DESCRIPTION

[0027] The following, in conjunction with the accompanying drawings, further describes the technical solution of the multidimensional clustering topic evolution model for network big data text parsing provided by the present application, so that technical personnel in the field can better understand the present application and implement it.

[0028] With the development of network information technology, people have to face the information confusion caused by information explosion while enjoying the convenience brought by information data. Text information is used in different scenarios as a carrier of information dissemination. Therefore, the text parsing and classification of text information has become a hot topic in the field of natural language processing.

[0029] Different from the previous vector space model representation, the topic model describes the documents in the corpus as a distribution about topics, and topics are distributions about terms, that is, a document is described by a fixed number of topics, and each topic is described by specific words. When faced with text parsing problems, which terms can express the text well often has a certain ambiguity; when faced with text classes, the spatial dimensions describing the text are generally very high. In text parsing and classification, the topic dimension of the topic model is crucial. The more topic dimensions there are, the higher the difficulty and computational complexity of classification. Fewer topic dimensions will discard a lot of important information. This application selects different topic dimensions and performs text parsing and text classification through the evaluation indicators of doubt degree and log-likelihood function, and concludes that the modeling effect based on the log-likelihood dynamic function is better than the modeling effect based on doubt degree.

[0030] (1) Estimation of the optimal topic dimension of the topic evolution model. This application combines the language model evaluation index with the structural characteristics of the data evolution model itself, and adopts a method based on doubt degree and log-likelihood dynamic function to judge the topic dimension in the corpus, calculates the changes between doubt degree and log-likelihood dynamic function under different topic dimensions, and uses M-layer cross-validation to find the topic dimension with the lowest doubt degree and the topic dimension with the largest log-likelihood dynamic function, thereby obtaining different optimal topic dimensions; (2) Topic evolution model based on text analysis. This application uses visualization methods to evaluate the effect of modeling under different topic dimensions from the perspective of text analysis. By highlighting the analysis method of high-frequency words in the text analysis of "document-term", a text analysis method based on data evolution and a term analysis method in the topic space are constructed. At the same time, the optimal topic dimension is determined by the distribution of "topic-topic" and "topic-term" after modeling; (3) Topic evolution model based on text classification. Based on the distribution relationship between documents and topics, text representation is constructed through topic distribution and used as feature vector. Different topics are modeled through text classification comparison to determine the optimal topic dimension. At the same time, a comparative experiment is conducted with the traditional vector space model. The feasibility and effectiveness of text classification based on topic dimensionality reduction are proved through experiments.

[0031] 1. Optimal Topic Dimension Estimation for Data Evolution Model The first difficulty in establishing a data evolution model is to estimate how many topics M there are in the corpus, and the topic dimension of the data evolution model cannot be known in advance. This application constructs a strategy for determining the topic dimension in the data evolution model based on the change in doubt and the log-likelihood dynamic function. The doubt gradually decreases and tends to be flat as the number of topic dimensions increases, while the log-likelihood dynamic function gradually increases and tends to be flat as the number of topic dimensions increases. As long as the topic dimension is large enough, the doubt must be low enough or the log-likelihood dynamic function is large enough, but such a model has the problem of overfitting. By setting different topics M, observing the current number of topics M, using multi-layer cross-validation doubt and log-likelihood dynamic function, and determining the M value by comparing the visualization graph, which is the optimal topic dimension estimate of the current corpus. The overall block diagram of the optimal topic dimension estimation of the data evolution model is shown in the figure below. Figure 1 .

[0032] 1. Data Evolution Model Assume that the test text set is represented by M, the terms in the bag-of-words model are represented by w, and the log-likelihood value describes the distribution of the word in the topic space. First, the prior parameters of the "document-term" matrix are given, and the prior distribution parameter values ​​are determined by using Gibbs sampling. The corpus is divided into a training set and a test set, and the model parameters are trained using the training set. Then, the log-likelihood dynamic function of the test set is calculated using VEM estimation. Estimating the topic dimension is a process of continuous model comparison. It analyzes the changes in the mean of the model's doubt and the mean of the log-likelihood dynamic function when the topic dimension changes. The cross-validation method is used to test the data. When modeling the corpus, the text data is randomly divided into M groups of equal size, one group is selected as the test set, and the remaining M-1 groups of data are modeled to train the topic evolution model. The trained topic evolution model is then used to model the test set, and a set of doubts and log-likelihood dynamic functions are calculated. Under the same conditions, this process is repeated many times to obtain multiple sets of doubts and log-likelihood dynamic functions, and the mean of multiple sets of doubts and log-likelihood dynamic functions is output as the result to complete multi-layer cross-validation and avoid overfitting or underfitting of the model.

[0033] When evaluating the modeling effect of the topic evolution model, it is determined that the lower the doubt, the better, and the larger the log-likelihood dynamic function, the better. After a topic modeling, the doubt and log-likelihood dynamic function are calculated at the same time, and a multi-layer cross-check is performed under the current model parameters to obtain the average doubt and log-likelihood dynamic function. Then, the doubt and log-likelihood dynamic function trends output by the topic dimension analysis are changed. The process of solving the average doubt and log-likelihood dynamic function is as follows: Step 1: Given the subject dimension; Step 2: Select the Gibbs topic modeling algorithm; Step 3: Solve the current modeling doubt and log-likelihood dynamic function; Step 4: Repeat the M-layer cross-validation six times and record the average doubt and log-likelihood dynamic function.

[0034] The topic dimension of the topic evolution model is simulated and determined through the M-layer cross-validation method. At the same time, the two evaluation indicators of doubt degree and log-likelihood function value are combined to determine the topic value, so as to make up for the simple reliance on doubt degree to evaluate the quality of the model.

[0035] 2. Topic Evolution Model Based on Text Parsing In order to solve the problem of how to visualize and analyze texts in a corpus, this application analyzes the texts after topic modeling and evaluates the effect of modeling to obtain the optimal modeling topic dimension. First, a topic evolution model is constructed based on the estimated topic dimension corpus modeling to express two sets of "document-topic" and "topic-term" relationships. One is to express the core of the corpus by extracting terms under the topic as keywords, and to give an evaluation through comparison based on the analysis of the text word frequency word cloud; the other is to use the topic similarity evaluation method between topic structures as an evaluation of topic modeling, and to evaluate the modeling effects of different topic dimensions from an intuitive and objective perspective from the data evolution vis visualization analysis. Figure 2 shown.

[0036] 1. Word cloud analysis based on text word frequency In word cloud analysis, the weight of terms is measured based on the frequency of words in the text, and high-weight terms are output as the subject words of the word cloud. The process of generating a word cloud includes two steps: first, the weight of each word is calculated for the preprocessed text or corpus; second, the size of the word in the composition space is estimated according to the size of the word weight and the word cloud is output. When outputting the word cloud, the priority of terms with different weights in the composition space is considered, and the stability and rationality of the term layout are met by a compact layout method.

[0037] 2. Text parsing based on word cloud The corpus is modeled through the data evolution model, and the distribution of "topic-term" in the corpus is learned. According to the estimated numbers of two topics, the corpus is modeled with a topic dimension of 20 and a topic dimension of 60, respectively, to form two groups of comparisons. The distribution of terms under each topic is analyzed in the modeled corpus. The terms under each topic are distributed according to the probability of p(w|z). Words with high probability are prioritized in each topic space. It is approximately determined that words with high frequency in different topic spaces are prioritized in front. In the two groups of modeling, 120 words with higher probability in the topic space are extracted respectively and the word frequency is counted, and then displayed in the form of word cloud.

[0038] 3. Visualization analysis based on the topic evolution model Through data evolution vis network visualization as a text parsing tool after topic modeling of different topic dimensions, through multi-scale analysis, the relative distance between objects is grasped, and the expression of data in low-dimensional space is obtained. The distribution of corpus documents in the topic space and the distribution of terms in the topic space after data evolution modeling are analyzed, and the probability size information of terms is expressed by the ranking of terms.

[0039] After preprocessing the text in the corpus, the text content of text_content is formed, and then the text is subject modeled on this basis. The modeling results are visualized through the topic evolution model. The output results are formulated in the form of web pages by writing scripts. The algorithm process of the modeling process with a topic dimension of 20 is as follows: (1) text_content# indicates the text content after word segmentation; (2) Convert text_content into list type and assign it to text_contents; (3) Remove the empty items in text_contents and assign them to doc.list; (4) Convert doc.list to table type and assign it to term.table; (5) Sort term.table in descending order; (6) Get the header of term.table; (7) Specify the del deletion parameter to be a word frequency less than 5; (8) Get the header of term.table; (9) Remove the table items in term.table that satisfy del; (10) Assign the names in term.table to vocab; (11) Specify the method get.terms with the parameter x; (12) Get the position of index x in vocab; (13) index cannot be empty; (14) Combine the position of index and the number of times x appears into a vector; (15) Assign the result of executing the get.terms method on doc.list to documents; (16) Save the results to fit; (17) Calculate the probability distribution theta of the documents in the dataset; (18) Label the result as theta_20; (19) Calculate the probability distribution of terms in the topic space; (20) Mark the result as phi_20; (21) Convert the topic terms into a list topic_words; (22) Display topic_words; (23) Set term weights term.weights; (24) Get the length of the document doc.length; (25) Generate json variables in web page information; (26) Get the document length doc.length and vocabulary vocab; (27) Set term weights term.weights (28) Set json output serVis; The ranking of terms is changed by the adjustable parameter a to help explain the topic. If a is close to 1, the terms at the top are those with higher frequency in the topic space; if a is closer to 0, the terms at the top are those with more unique characteristics in the topic space. By adjusting the size of a, the weight of the terms under the topic is changed, affecting the order of the terms. The value of a is determined by the following formula: Formula 1 Based on the relevance of a certain word topic, it is adjusted by a. By comparing the frequency of a given word in the corpus with the topic-specific frequency, it is quickly determined whether it is a highly relevant selected topic; θ represents the vector of topic distribution, p(θ) is the distribution of θ, N represents the number of words in the corpus, w_n is the nth word w, z_n represents the selected topic, p(z|θ) represents the probability of topic z under the topic distribution θ, p(w|z) represents the probability of word w under topic z, expressed as an M×V matrix, M represents the topic, V represents the word in the corpus, α is the parameter describing the distribution p(θ), affecting the distribution of the topic θ vector, β is the parameter describing the distribution p(w|z), affecting the distribution of the word under topic z; The relationship between topics is analyzed through multidimensional scaling. The high-dimensional topic distribution is projected into two-dimensional space, and the size and distance of the bubble chart are used to describe the differences between topics. The modeling effect is better when M=20.

[0040] From the perspective of the distribution of terms under a topic, the fewer topic structures a term has associated with it in the topic space, the more evenly distributed the "topic-term" is. The modeling effect is evaluated by the relationship between different modeling topic dimensions and the topic similarities obtained after modeling. That is, when the average similarity between topics is the smallest, the model is optimal.

[0041] A large number of experiments have shown that there is an optimal structure in the topic structure implicit in the extracted data set by performing data evolution topic modeling on the corpus. Therefore, the probability distribution of topics in the term space is used to represent the topic vector, the cosine distance is used to represent the difference of the topic, and cosine(z i ,z j ) is smaller, the greater the difference in the subject space: Formula 2 The quality of the topic structure is characterized by the average similarity of the topic vectors: Formula 3 The modeling effects of the two are compared directly by comparing the average topic similarities under different topic dimensions. The topic similarities under different topic dimensions are: M=20, Avg_cosine=0.0283, M=60, Avg_cosine=0.03008073; By comparing the average similarities of all topics in the topic evolution model under different topic dimensions, it is determined that when the topic dimension is 20, the difference of topic vectors in the topic space is greater, and it is determined that the modeling performance when the topic dimension is 20 is relatively better.

[0042] 3. Topic Evolution Model Based on Multi-dimensional Text Clustering The "document-topic" matrix obtained after data evolution modeling of the test set in the corpus is used as the training set of the nearest neighbor classifier. The test set is obtained in the same way and predicted by the nearest neighbor classifier. First, the text representation of data evolution modeling is constructed by estimating different topic dimensions, and the advantages and disadvantages of different topic dimensions are estimated through classification effect feedback; then the text representation of the optimal data evolution modeling is compared with the text representation constructed by the vector space model to verify the feasibility and effectiveness of the data evolution model in classification, such as Figure 3 shown.

[0043] 1. Text classification model based on data evolution dimensionality reduction A relatively important step in the text classification problem is to represent the text as a vector. Generally, the obvious problem when using a vector representation based on terms is that there is often a huge spatial dimension. Feature extraction is used to reduce the spatial dimension and thus reduce the dimension. The reason for dimensionality reduction is due to two considerations: the first is to facilitate calculation, and excessively high dimensions represent time-consuming calculations; the second is to improve accuracy, and in the process of dimensionality reduction, useless terms are removed and more useful features are retained. Regardless of which feature selection algorithm is used, the feature vector that finally represents the text is a part of the original term set, and the representation of the text still only stays at the word level, and there is no in-depth study of the text semantics. The present application constructs a method for classifying text by using a data evolution model to represent the text after modeling the corpus.

[0044] like Figure 4 As shown, the specific steps are as follows: Step 1: Use the data evolution model to model the training document set and obtain the distribution of documents on topics as the vector representation of the documents; Step 2: Use the nearest neighbor algorithm to train the vector in step 1; Step 3: Use the results in step 1 to model the test document set and obtain the topic distribution in the test document set; Step 4: Represent the result as a text vector with a fixed number of text features; Step 5: Predict the category of the text vector using the classifier obtained in step 2.

[0045] 2. Text classification model based on the nearest neighbor By representing the text in the corpus as a feature vector, the category of the unknown text is determined to be the same as the majority category of T adjacent samples in the feature item space in the classification. The number of the most adjacent samples T and the distance function are known. The T value is the square root of the number of training sets, and the distance function is the Euclidean distance.

[0046] like Figure 5 As shown, the algorithm flow is as follows: Step 1: First, divide the corpus into a test document set and a training document set, and perform preprocessing; Step 2: Extract specific parts of speech for all document sets in the corpus and convert them into text representation based on word vectors; Step 3: Select consistent feature vectors based on text word frequency; Step 4: For the vector to be classified in the training document set, traverse the distance between the vector and the known classification vector; Step 5: Select the nearest vectors in T vector spaces and calculate the category weights of the majority vectors; Step 6: Label the unknown vector as the class with the largest weight for the majority class.

Claims

1. A multi-dimensional clustering topic evolution model for network big data text analysis, characterized by: Construct an evaluation method for estimating topic dimensions in data evolution models from the perspective of text parsing and text classification. By calculating the dynamic functions of doubt and log-likelihood under different topic dimensions, the topic dimensions are estimated in a data-driven way. Then, based on the results of corpus modeling using the estimated topic dimension, the optimal topic dimension is judged in combination with text analysis and text classification. Through visualization under the topic evolution model, the text content is intuitively displayed in the form of word clouds and bubble charts from the perspectives of word frequency and document, word frequency and classification, and topic distribution based on data evolution, and a dynamic comparison of the modeling effects based on the log-likelihood dynamic function and the modeling effect based on the degree of doubt is obtained. M1-Estimation of the optimal topic dimension of the topic evolution model: Combining the evaluation indicators of the language model with the structural characteristics of the data evolution model itself, a method based on doubt and log-likelihood dynamic function is used to judge the topic dimension in the corpus, and the changes between doubt and log-likelihood dynamic function under different topic dimensions are calculated. Through M-layer cross-validation, the topic dimension with the lowest doubt and the topic dimension with the largest log-likelihood dynamic function are obtained, thereby obtaining different optimal topic dimensions; M2-Topic evolution model based on text analysis: From the perspective of text analysis, the modeling effect under different topic dimensions is evaluated by using visualization methods. By highlighting the analysis method of high-frequency words in the text analysis of "document-term", a text analysis method based on data evolution and a term analysis method in the topic space are constructed. At the same time, the optimal topic dimension is determined by the distribution of "topic-topic" and "topic-term" after modeling; M3-Topic evolution model based on text classification: Based on the distribution relationship between documents and topics, text representation is constructed through topic distribution and used as feature vector. Different topics are modeled through text classification comparison to determine the optimal topic dimension. The topic evolution model is optimized through the semantic expression of high-probability terms and specific comparison results, and the modeling effect of the topic dimension is evaluated based on different indicators.

2. According to the network big data text analysis multi-dimensional clustering topic evolution model described in claim 1, it is characterized by: Optimal topic dimension estimation of the data evolution model: By setting different topics M, observing the current number of topics M, using multi-layer cross-validation doubt and log-likelihood dynamic functions, and determining the M value by comparing the visualization graph, which is the optimal topic dimension estimate of the current corpus.

3. According to the network big data text analysis multi-dimensional clustering topic evolution model described in claim 1, it is characterized by: Data evolution model: Assume that the test text set is represented by M, the terms in the bag-of-words model are represented by w, and the log-likelihood value describes the distribution of the word in the topic space. First, the prior parameters of the "document-term" matrix are given, and the prior distribution parameter values ​​are determined by using Gibbs sampling. The corpus is divided into a training set and a test set, and the model parameters are trained using the training set. Then, the dynamic log-likelihood function of the test set is calculated using VEM estimation. Estimating the topic dimension is a process of continuous model comparison. It analyzes the changes in the mean of the model's doubt and the mean of the log-likelihood dynamic function when the topic dimension changes. The cross-validation method is used to test the data. When modeling the corpus, the text data is randomly divided into M groups of equal size. One group is selected as the test set. The remaining M-1 groups of data are modeled to train the topic evolution model. The trained topic evolution model is then used to model the test set. A set of doubt and log-likelihood dynamic functions are calculated. Repeating this process multiple times under the same conditions will obtain multiple sets of doubt and log-likelihood dynamic functions, and the mean of multiple sets of doubt and log-likelihood dynamic functions is output as the result to complete a multi-layer cross-validation. When evaluating the modeling effect of the topic evolution model, it is determined that the lower the doubt, the better, and the larger the log-likelihood dynamic function, the better. After a topic modeling, the doubt and log-likelihood dynamic function are calculated at the same time. A multi-layer cross-check is performed under the current model parameters to obtain the average doubt and log-likelihood dynamic function. Then, the doubt and log-likelihood dynamic function trends output by the topic dimension analysis are changed. The process of solving the average doubt and log-likelihood dynamic function is as follows: Step 1: Given the subject dimension; Step 2: Select the Gibbs topic modeling algorithm; Step 3: Solve the current modeling doubt and log-likelihood dynamic function; Step 4: Repeat the M-layer cross-validation 6 times and record the average doubt and log-likelihood dynamic function; The topic dimension of the topic evolution model is simulated and determined through the M-layer cross-validation method, and the topic value is determined by combining the two evaluation indicators of doubt degree and log-likelihood function value.

4. According to claim 1, the multi-dimensional clustering topic evolution model for network big data text analysis is characterized in that: Topic evolution model based on text analysis: By analyzing the text after topic modeling and evaluating the effect of modeling, the optimal modeling topic dimension is obtained. First, a topic evolution model is constructed based on the estimated topic dimension corpus modeling to express two sets of "document-topic" and "topic-term" relationships. One is to express the core of the corpus by extracting terms under the topic as keywords, and give an evaluation through comparison based on the analysis of text word frequency word cloud; the other is to use the topic similarity evaluation method between topic structures as an evaluation of topic modeling, and evaluate the modeling effects of different topic dimensions from an intuitive and objective perspective from the data evolution vis visualization analysis.

5. According to claim 1, the multi-dimensional clustering topic evolution model for network big data text analysis is characterized in that: Word cloud analysis based on text word frequency: In word cloud analysis, the weight of words is measured based on the word frequency of the text, and the words with high weight are output as the subject words of the word cloud. The process of generating a word cloud includes two steps: first, the weight of each word is calculated for the preprocessed text or corpus; Secondly, the word cloud is output based on the estimated size of the word weight in the composition space. When outputting the word cloud, the priority of words with different weights in the composition space is considered, and the stability and rationality of the word layout are met by using a compact layout method.

6. According to claim 1, the multi-dimensional clustering topic evolution model for network big data text analysis is characterized in that: Text parsing based on word cloud: The corpus is modeled through the data evolution model to learn the distribution of "topic-term" in the corpus. According to the estimated numbers of two topics, the corpus is modeled with a topic dimension of 20 and a topic dimension of 60, respectively, to form two groups of comparisons. The distribution of terms under each topic is analyzed in the modeled corpus. The terms under each topic are distributed according to the probability of p(w|z). Words with high probability are prioritized in each topic space. It is approximately determined that words with high frequency in different topic spaces are prioritized in front. In the two groups of modeling, 120 words with high probability in the topic space are extracted and the word frequency is counted, and then displayed in the form of word cloud.

7. According to claim 1, the multi-dimensional clustering theme evolution model for network big data text analysis is characterized in that: Visual analysis based on the topic evolution model: Through data evolution vis network visualization as a text parsing tool after topic modeling of different topic dimensions, through multi-scale analysis, the relative distance between objects is grasped, and the expression form of data in low-dimensional space is obtained. The distribution of corpus documents in the topic space and the distribution of terms in the topic space after data evolution modeling are analyzed, and the probability size information of terms is expressed by the ranking of terms. After preprocessing the text in the corpus, the text content of text_content is formed, and then the text is subject modeled on this basis. The modeling results are visualized through the topic evolution model, and the output results are formulated in the form of web pages by writing scripts; The ranking of terms is changed by the adjustable parameter a to help explain the topic. If a is close to 1, the terms at the top are those with higher frequency in the topic space; if a is closer to 0, the terms at the top are those with more unique characteristics in the topic space. By adjusting the size of a, the weight of the terms under the topic is changed, affecting the order of the terms. The value of a is determined by the following formula: Formula 1 Based on the relevance of a certain word topic, it is adjusted by a. By comparing the frequency of a given word in the corpus with the topic-specific frequency, it is quickly determined whether it is a highly relevant selected topic; θ represents the vector of topic distribution, p(θ) is the distribution of θ, N represents the number of words in the corpus, w_n is the nth word w, z_n represents the selected topic, p(z|θ) represents the probability of topic z under the topic distribution θ, p(w|z) represents the probability of word w under topic z, expressed as an M×V matrix, M represents the topic, V represents the word in the corpus, α is the parameter describing the distribution p(θ), affecting the distribution of the topic θ vector, β is the parameter describing the distribution p(w|z), affecting the distribution of the word under topic z; The relationship between topics is analyzed by multidimensional scaling, which projects the high-dimensional topic distribution into two-dimensional space, and describes the differences between topics by the size and distance of the bubble chart; From the perspective of the distribution of terms under a topic, the fewer topic structures a term has associated with it in the topic space, the more evenly the "topic-term" distribution is. The modeling effect is evaluated by the relationship between different modeling topic dimensions and the topic similarity obtained after modeling. That is, when the average similarity between topics is the smallest, the model is optimal. The probability distribution of topics in the term space is used to represent the topic vector, the cosine distance is used to represent the difference of topics, and cosine(z i ,z j ) is smaller, the greater the difference in the subject space: Formula 2 The quality of the topic structure is characterized by the average similarity of the topic vectors: Formula 3 The modeling effects of the two are compared directly by comparing the average topic similarities under different topic dimension modeling, and the average similarities of all topics in the topic evolution model under different topic dimensions are compared. It is determined that when the topic dimension is 20, the difference of topic vectors in the topic space is greater, and it is determined that the modeling performance with a topic dimension of 20 is relatively better.

8. According to the network big data text analysis multi-dimensional clustering topic evolution model described in claim 1, it is characterized by: Topic evolution model based on text multi-dimensional clustering: The "document-topic" matrix obtained after data evolution modeling of the test set in the corpus is used as the training set of the nearest neighbor classifier. The test set is obtained in the same way and predictions are made through the nearest neighbor classifier. First, the text representation of data evolution modeling is constructed through estimation of different topic dimensions, and the pros and cons of different topic dimensions are estimated through classification effect feedback; then the optimal text representation of data evolution modeling is compared with the text representation constructed by the vector space model.

9. According to claim 1, the multi-dimensional clustering topic evolution model for network big data text analysis is characterized in that: Text classification model based on data evolution dimensionality reduction: Construct a method for text classification by using the text representation after modeling the corpus using the data evolution model. The specific steps are as follows: Step 1: Use the data evolution model to model the training document set and obtain the distribution of documents on topics as the vector representation of the documents; Step 2: Use the nearest neighbor algorithm to train the vector in step 1; Step 3: Use the results in step 1 to model the test document set and obtain the topic distribution in the test document set; Step 4: Represent the result as a text vector with a fixed number of text features; Step 5: Predict the category of the text vector using the classifier obtained in step 2.

10. According to claim 1, the multi-dimensional clustering topic evolution model for network big data text analysis is characterized in that: Based on the nearest neighbor text classification model: By representing the text in the corpus as a feature vector, the category of the unknown text is determined to be the same as the majority of the categories of the T adjacent samples in the feature item space. The number of nearest neighbor samples T and the distance function are known. The T value is the square root of the number of training sets, and the distance function is the Euclidean distance. The algorithm flow is as follows: Step 1: First, divide the corpus into a test document set and a training document set, and perform preprocessing; Step 2: Extract specific parts of speech for all document sets in the corpus and convert them into text representation based on word vectors; Step 3: Select consistent feature vectors based on text word frequency; Step 4: For the vector to be classified in the training document set, traverse the distance between the vector and the known classification vector; Step 5: Select the nearest vectors in T vector spaces and calculate the category weights of the majority vectors; Step 6: Label the unknown vector as the class with the largest weight for the majority class.

Images