TWI659321B - System and method for analyzing industry relevance - Google Patents

Abstract

The invention discloses an industry correlation analysis system and method. Such an industry correlation analysis system includes a user interface, a server, a memory, and a processor. This method of industry relevance analysis includes: searching for multiple related words based on keywords; searching for multiple news related to the related words within a preset time period, and calculating for each related word based on the news. And generate a word cloud, where each word cloud includes multiple text features; comparing the text features to calculate the overlap rate of each text feature, and sorting the text features according to the overlap rate from large to small; and extracting A number of text features with a large overlap rate are generated according to the extracted text features and the relationship between the extracted text features and the related words.

Description

Industrial relevance analysis system and method

The present invention relates to an industry correlation analysis system and method, and particularly to an industry correlation analysis system and method capable of performing correlation analysis for multiple companies in different industries.

For industry researchers, it is often necessary to look for key topics from recent news on various industries and recent articles on various industries that have been discussed on social media. In addition, recent news from various industries and recent discussions on various industries on social media often reveal differences or similarities between companies in the same industry, between companies, or between different industries.

However, due to the large amount of information, key issues must be compiled from the recent news of various industries and recent articles on various industries that have been discussed on social media. It is also necessary to further unify companies or companies in the same industry or between companies. Differences or similarities between industries are not easy. If we can effectively summarize the key topics from the recent news of various industries and recent articles on various industries on social media, and further unify the differences between companies in the same industry or between companies or between different industries, or Similarly, industry researchers have the opportunity to judge the characteristics, transformation opportunities, or potential crises of a company or an industry based on current events.

The industry correlation analysis system provided by the present invention includes a user interface, a server, a memory and a processor. User interface settings to enter keywords and display analysis results. The server is configured to run at least one database. Memory settings to store analysis programs. The processor is connected to the user interface, the server, and the memory, and is configured to execute an analysis program to perform the following operations: based on keywords, the processor searches for a plurality of related words in a database; the processor searches for a pre- Set a number of news related to the related words in the time section, and calculate and generate a word cloud for each related word based on the news, where each word cloud includes multiple text features; compare the word clouds The text features to calculate an overlap rate of each text feature, and sort the text features of the word cloud from large to small according to the overlap rate; and take out several text features with a large overlap rate, and A correlation analysis diagram is generated according to the extracted text features and the relationship between the extracted text features and the related words as an analysis result.

In an embodiment of the industry correlation analysis system provided by the present invention, the processor further executes an analysis program to further perform the following operations: extracting several text features with a small overlap rate, and according to the extracted text features and The relationship between the extracted text features and the related words generates another correlation analysis diagram as an analysis result. In another embodiment of the industry correlation analysis system provided by the present invention, the processor further executes an analysis program to further perform the following operations: extracting several text features with a large overlap rate and several text features with a small overlap rate , And generate another correlation analysis diagram according to the extracted text features and the relationship between the extracted text features and the related words as an analysis result.

In addition, the industrial correlation analysis method provided by the present invention is applicable to the aforementioned industrial correlation analysis system. The industrial relevance analysis method provided by the present invention includes: searching for a plurality of related words in a database according to keywords; searching for a plurality of news related to the related words in a preset time section in the database, and According to the news, a word cloud is calculated and generated for each associated word, where each word cloud includes multiple text features; the word features of the word clouds are compared to calculate each An overlap rate of the text features, and sort the text features of the word cloud in descending order according to the overlap rate; and extract several text features with a large overlap rate, and according to the extracted text features and The relationship between the extracted text features and the related words generates a correlation analysis diagram as an analysis result.

In an embodiment of the industry association analysis method provided by the present invention, the industry association analysis method further includes: extracting several text features with a small overlap ratio, and according to the text features being retrieved and the extracted ones. The relationship between the text features and the related words generates another correlation analysis diagram as the analysis result. In another embodiment of the industry association analysis method provided by the present invention, the industry association analysis method further includes: taking out several text features with a large overlap rate and several text features with a small overlap rate, and The extracted text features and the relationship between the extracted text features and the related words generate another correlation analysis diagram as an analysis result.

Through the industry correlation analysis system and method provided by the present invention, after generating multiple word clouds based on keywords and related news reports, these word clouds can be analyzed and the word clouds can be displayed in the analysis results. Relevance or difference. In this way, it will help industry researchers to analyze and compile a large number of news information.

10‧‧‧ processor

11‧‧‧user interface

12‧‧‧Server

13‧‧‧Database

14‧‧‧Memory

15‧‧‧analysis program

S210, S220, S230, S240 ‧‧‧ steps

S201, S202, S203, S204, S205, S206‧‧‧ steps

S242, S244‧‧‧ steps

FIG. 1 is a block diagram of an industry correlation analysis system according to an exemplary embodiment of the present invention.

FIG. 2 is a block diagram of an industrial correlation analysis method according to an exemplary embodiment of the present invention.

FIG. 3 is a block diagram illustrating a gold industry correlation analysis method according to another exemplary embodiment of the present invention.

FIG. 4A is a schematic diagram of a correlation analysis diagram according to an exemplary embodiment of the present invention.

FIG. 4B is a schematic diagram of a correlation analysis diagram according to another exemplary embodiment of the present invention.

FIG. 4C is a schematic diagram of a correlation analysis diagram according to another exemplary embodiment of the present invention.

Various exemplary embodiments will be described more fully hereinafter with reference to the accompanying drawings, in which some exemplary embodiments are shown. However, the inventive concept may be embodied in many different forms and should not be construed as limited to the exemplary embodiments set forth herein. Rather, these exemplary embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the inventive concept to those skilled in the art. In the drawings, similar numbers always indicate similar elements.

In general, in order to be able to analyze, aggregate and generate analysis results with a large amount of news information, the industry correlation analysis system and method provided by the present invention generate multiple word clouds based on a large amount of news , And further analyze and aggregate these word clouds to generate a correlation analysis diagram that is clear to users. In the following, several embodiments will be used to explain the industry correlation analysis system and method provided by the present invention.

[An embodiment of the industry correlation analysis system]

The structure of the industry correlation analysis system of the present invention is described first. Please refer to FIG. 1. FIG. 1 is a block diagram of the industry correlation analysis system according to an exemplary embodiment of the present invention.

As shown in FIG. 1, the industry correlation analysis system provided in this embodiment includes a processor 10, a user interface 11, a server 12, and a memory 14. The user interface 11 is configured to input keywords and display analysis results. The server 12 is configured to run at least one database 13. The memory 14 is configured to store an analysis program 15. The processor 10 is connected to the user interface 11 and the server 12 and the memory 14. The processor 10, the user interface 11 and the memory 14 in the financial unstructured text analysis system provided in this embodiment may be Electronic devices, such as personal computers, smart phones, etc. The server 12 in the financial unstructured text analysis system provided in this embodiment may be implemented by a server device capable of performing network communication with an electronic device.

[An embodiment of the analysis method of industrial correlation]

Please refer to FIG. 2, which is a block diagram of an industrial correlation analysis method according to an exemplary embodiment of the present invention.

The industry association analysis method provided in this embodiment is implemented by the processor 10 in the industry association analysis system shown in FIG. 1 executing an analysis program 15 stored in the memory 14, so please refer to FIG. 1 at the same time. With Figure 2 to facilitate understanding. As shown in FIG. 2, generally speaking, the industrial correlation analysis method provided in this embodiment includes the following steps: searching for a plurality of related words in a database according to keywords (step S210); Set a number of news related to the related words in the time section, and calculate and generate a word cloud for each related word based on the news, where each word cloud includes multiple text features (step S220); compare The text features of the word clouds to calculate an overlap rate of each text feature, and sort the text features of the word clouds from large to small according to the overlap rate (step S230); and extract the overlap rate A larger number of text features are generated as a result of analysis based on the extracted text features and the relationship between the extracted text features and the related words as an analysis result (step S240).

Next, the details of each step in the industrial correlation analysis method provided in this embodiment are explained.

In step S210, when the user inputs a keyword through the user interface 11, according to the keyword, the processor 10 searches the database 13 for a plurality of related words through the server 12. In this embodiment, the keyword is an industry name, and the related words are multiple company names, but the invention is not limited thereto. For example, if the user inputs "display" through the user interface 11, the processor 10 searches the database 13 through the server 12 for a plurality of related words corresponding to the keyword "display", such as: "A company "," Company B "," Company C ".

Then, in step S220, the processor 10 searches the database 13 through the server 12 for a plurality of news related to each associated word within a preset time period. In this embodiment, the server 12 runs at least one database 13, and the data source of the database 13 may be, for example, news released by major news networks. In addition, the aforementioned multiple news related to each related word are the news in which the related word exists in the content. Assuming that the related words are "Company A", "Company B", and "Company C", the processor 10 will be based on multiple news related to "Company A", multiple news related to "Company B", and A number of news stories related to "Company C" were calculated and generated as a word cloud. It should be noted that each word cloud generated in step S220 includes multiple word features, and the word features referred to herein are multiple words used to form each word cloud.

Next, in step S230, the processor 10 compares the text features in the multiple word clouds to calculate an overlap rate of each text feature, and calculates the text features of the word clouds according to the overlap rate. Sort from largest to smallest.

Simply put, the overlap rate is the probability of a text feature appearing in those word clouds. The calculation of the overlap rate depends first on the appearance of a word feature in several word clouds, and also on the number of times that the word feature appears in the word clouds. It is assumed that, in step S220, word cloud I, word cloud II, and word cloud III are generated for related words "company A", "company B", and "company C". At this time, if a text feature appears in the word cloud I, word cloud II, and word cloud III at the same time, and the text feature appears in the word clouds multiple times, the text feature has a large overlap rate; on the contrary, If a text feature only appears in word cloud I, word cloud II, or word cloud III, and the text feature appears in the word cloud only once or twice, then the text feature has a small overlap rate.

Finally, in step S240, the processor 10 extracts several text features with a large overlap ratio, and generates an association analysis based on the extracted text features and the relationship between the extracted text features and the related words. Figure as the analysis result.

For the correlation analysis diagram generated in step S240, please refer to FIG. 4A. FIG. 4A is a schematic diagram of a correlation analysis diagram according to an exemplary embodiment of the present invention. Assume that in step S240, ten text features with a large overlap rate need to be extracted, and these ten overlaps The text features with larger rates are `` Company D '', `` Company E '', `` Foreign Investment '', `` Shareholding '', `` Glass '', `` Panel '', `` Order '', `` OEM '', `` Lens '' and `` Mobile Phone '' ". At this time, an example of the correlation analysis diagram generated in step S240 is shown in FIG. 4A.

It can be seen from FIG. 4A that among the ten text features, “glass”, “foundry”, and “E company” are the text features with a large overlap rate, and “glass”, “foundry”, and “E company” are The association between company A, company B and company C. In addition, compared with "glass", "foundry", "company E", although the overlap rate of other text features is small, these text features can be regarded as company A and company B, company A and company C or Association between Company B and Company C.

That is, according to the analysis result (ie, the correlation analysis diagram) generated by the industrial correlation analysis method provided in this embodiment, the correlation between multiple companies in an industry can be clearly understood.

[Another embodiment of the industrial correlation analysis method]

Please refer to FIG. 3, which is a block diagram of an industrial correlation analysis method according to another exemplary embodiment of the present invention.

The industry association analysis method provided in this embodiment is implemented by the processor 10 in the industry association analysis system shown in FIG. 1 executing an analysis program 15 stored in the memory 14, so please refer to FIG. 1 at the same time. With Figure 3 to facilitate understanding. It should be noted that steps S220 to S240 in the industrial correlation analysis method provided in this embodiment are similar to steps S220 to S240 in the industrial correlation analysis method shown in FIG. 2, so in the following description, Only the differences between the two will be described. For details about steps S220 to S240, please refer to the foregoing embodiment.

One of the differences between the industry association analysis method provided in this embodiment and the industry association analysis method shown in FIG. 2 is that the industry association analysis method provided in this embodiment further includes steps S201 to S206.

In step S201, the processor 10 determines whether the keyword entered by the user is an industry name. In this embodiment, the database 13 also stores an industry-individual association table. The industry-individual association table includes multiple industry names and multiple company names. Called. In this industry-individual association table, each industry name corresponds to one or more company names, and each company name also corresponds to one or more industry names.

If the processor 10 determines that the keyword input by the user is an industry name, it proceeds to step S202. In step S202, the processor 10 searches for a plurality of related words according to the keywords entered by the user and the aforementioned industry and individual association table. For example, the keyword entered by the user is "display", and the related words searched through the industry and individual association table are "company A", "company B", and company "C".

On the other hand, if the processor 10 determines that the keyword input by the user is not an industry name, it proceeds to step S203. In step S203, the processor 10 further determines whether the keyword entered by the user is a company name. If the processor 10 determines that the keyword entered by the user is not a company name, it proceeds to step S204 to wait for the user to enter the next new keyword. On the other hand, if the processor 10 determines that the keyword input by the user is a company name, it proceeds to step S205.

In step S205, the processor 10 searches the database 13 for a plurality of auxiliary related words according to the keywords. For example, if the keyword entered by the user is "Company A", according to the aforementioned industry-individual association table, "Company A" corresponds to three industries: "display", "LED", and "mobile phone". In this step, the processor 10 searches for the three auxiliary related words of “display”, “LED” and “mobile phone”.

Finally, in step S206, the processor 10 searches the database 13 for a plurality of related words according to one of the auxiliary related words. Following the above example, the processor 10 searches the database 13 for a plurality of related words according to one of the auxiliary related words of “display”, “LED” and “mobile phone”. For example, the processor 10 searches for a plurality of related words “Company A”, “Company B”, and “Company C” according to the auxiliary related word “display”.

It is worth noting that in step S206, the user can select one of the auxiliary related words through the user interface 11, so that the processor 10 searches for a plurality of related words according to the selected auxiliary related words. In addition, if the user does not select one of the auxiliary related words through the user interface 11, the processor 10 will sort according to a preset Or randomly select one of the auxiliary related words as a basis to search for a plurality of related words. For example, the default ordering of auxiliary related words of the keyword "company A" may be "display" → "LED" → "mobile phone". According to the design of the analysis program 15, the processor 10 may search for a plurality of related words (where N is an integer greater than or equal to 1) according to the auxiliary related words ranked as the Nth rank.

Therefore, the processor 10 can perform steps S220 to S240 according to the plurality of related words generated in steps S202 and S206 to generate the correlation analysis diagram shown in FIG. 4A.

Another difference between the industry association analysis method provided in this embodiment and the industry association analysis method shown in FIG. 2 is that the industry association analysis method provided in this embodiment further includes steps S242 and S244.

In step S242, the processor 10 extracts several text features with a small overlap ratio, and generates a correlation analysis diagram according to the extracted text features and the relationship between the extracted text features and the related words. Take as analysis result.

For the correlation analysis diagram generated in step S242, please refer to FIG. 4B. FIG. 4B is a schematic diagram of a correlation analysis diagram according to another exemplary embodiment of the present invention. Assume that the plurality of related words generated according to step S202 and step S206 are "F company" and "G company", and it is assumed that in step S242, ten text features with a small overlap rate need to be extracted, and the ten overlap rates are small The text features are "H company", "petroleum", "supplier", "engineering", "plastic", "mold", "order", "lamp", "export" and "foundry". At this time, an example of the correlation analysis diagram generated in step S242 can be shown in FIG. 4B.

As can be seen from Figure 4B, among these ten text characteristics, the overlap rate of "H company", "oil", "supplier", "engineering", "lamps", "export" and "foundry" is very low or even Textual characteristics with overlapping rates approaching zero; that is, "Company H", "Petroleum", "Supplier", "Engineering", "Lighting", "Export" and "OEM" are F and G companies The difference is. In addition, although "plastic", "mold", and "order" are also textual features with low overlap, they are still the association between company F and company G.

In addition, in step S244, the processor 10 extracts a number of text features with a large overlap rate and a number of text features with a small overlap rate, and according to the extracted text features and the extracted text features and The relationship between these related words generates another correlation analysis diagram as the analysis result.

For the correlation analysis diagram generated in step S244, please refer to FIG. 4C. FIG. 4C is a schematic diagram of the correlation analysis diagram according to another exemplary embodiment of the present invention. Assume that the plurality of related words generated according to steps S202 and S206 are "company J", "K company", and "L company", and it is assumed that in step S242, five text features with a large overlap rate and ten overlaps need to be extracted The text features with lower overlap rate. The five text features with higher overlap rate are "Order", "Processor", "Mobile Phone", "Wafer" and "Display". The ten texts with lower overlap rate are The characteristics are "M Company", "Financial Report", "Media", "Land Capital", "Revenue", "Panel", "OEM", "Shareholders' Meeting", "Computer" and "Communication". At this time, an example of the correlation analysis diagram generated in step S244 is shown in FIG. 4C.

It can be seen from FIG. 4C that among these fifteen text features, “order”, “processor”, “mobile phone”, “wafer” and “display” are text features with a large overlap rate, among which “processor” and "Wafer" is the association between Company J, K and L, "Order" is the association between Company J and K, "Display" is the association between Company J and L, and "Mobile Phone" It is the association between company K and company L. On the other hand, "M Company", "Financial Report", "Media", "Land-owned", "Revenue", "Panel", "OEM", "Shareholders' Meeting", "Computer" and "Communication" overlap Rate is very low, or even the textual feature of overlapping rate is close to zero; that is, "M company", "financial report", media, "land capital", "revenue", "panel", "foundry", " "Shareholders' meeting", "Computer" and "Communication" are the differences between Company F and Company G.

According to the foregoing description, in addition to generating a correlation analysis diagram as shown in FIG. 4A according to several text features with a large overlap rate, to show the correlation of multiple companies in an industry, through the provided by this embodiment The industrial correlation analysis method can also generate the correlation analysis diagram shown in FIG. 4B according to several text features with a small overlap rate. To show the differences between multiple companies in an industry. Or, by using the industry correlation analysis method provided in this embodiment, a correlation analysis diagram as shown in FIG. 4C can be generated according to several text features with a large overlap rate and a small number, so as to simultaneously display the industry Relevance and differentiation of multiple companies.

It is worth noting that, for convenience of explanation, the multiple word cloud systems used to generate the correlation analysis diagram in the foregoing embodiments are respectively associated with multiple company names under the same industry name; however, the present invention is not limited thereto . In other embodiments, each industry name in the industry-individual association table may also correspond to multiple other industry names. In this case, when the keyword entered by the user through the user interface 11 is an industry name, according to the keyword, the processor 10 searches for a plurality of others corresponding to the industry name according to the industry-individual association table. Industry name. Then, the processor 10 generates a word cloud according to the industry name and other industry names corresponding thereto. Finally, the processor 10 generates a correlation analysis chart according to these word clouds to show the correlation and / or difference between different industries.

It is also worth noting that although the specific method is described with reference to the flowchart depicted in this document, those with ordinary knowledge in the technical field to which the invention belongs should easily understand that the industrial correlation analysis method provided by the invention The execution order of the steps in this step is not limited by this. That is to say, in the industrial correlation analysis method provided by other embodiments of the present invention, the execution order of each step may be changed, some steps may be combined, or some steps may be omitted.

[Possible effect of the embodiment]

In summary, through the industry correlation analysis system and method provided by the present invention, a correlation analysis map can be quickly generated based on the recent news of various industries and recent articles on various industries discussed on social media. To show the correlation and / or difference between different industries, or the correlation and / or difference between multiple companies in the same industry.

In other words, the industry correlation analysis system and method provided by the present invention can effectively carry out the industry's recent news and various social media The articles discussed summarize key issues, and further unify the differences or similarities between companies in the same industry, between companies, or between different industries.

Finally, it must be noted that, in the foregoing description, although the concept of the technology of the present invention has been specifically shown and described with a number of exemplary embodiments, those having ordinary knowledge in the field of this technology will understand that Various changes in form and detail may be made without departing from the scope of the concept of the technology of the present invention as defined by the following patent application scope.

Claims (10)

An industry correlation analysis system includes: a user interface configured to input a keyword and display an analysis result; a server configured to run at least one database; a memory configured to store an analysis program; and A processor connected to the user interface, the server and the memory, configured to execute the analysis program to perform the following operations: according to the keyword, the processor searches for a plurality of related words in the database; The processor searches the database for a plurality of news related to each related word within a preset time period, and calculates and generates a word cloud for each related word based on the news, where the word cloud includes multiple Text features; compare the text features of the word clouds to calculate an overlap rate related to the probability of each text feature appearing in the word clouds, and compare the text features of the word clouds according to the The overlapping rate is sorted from large to small; and several text features with a large overlap rate are taken out, and according to the extracted text features and the extracted text features, A relationship analysis graph is generated as a result of the analysis of the relationship with the related words, in which a text feature with a large overlap rate is a text feature appearing in the word clouds at the same time, and the text feature appears multiple times Word cloud, then the text feature has a large overlap rate. The industry correlation analysis system according to claim 1, wherein the processor executes the analysis program to further perform the following operations: extracting several text features with a small overlap rate, and according to the extracted text features and the The relationship between the extracted text features and the related words is used to generate another correlation analysis diagram as the analysis result. The industry correlation analysis system according to claim 1, wherein the processor executes the analysis program to further perform the following operations: extracting the text features with a large overlap rate and the text features with a small overlap rate, According to the extracted text features and the relationship between the extracted text features and the related words, another correlation analysis diagram is generated as the analysis result. The industry correlation analysis system according to claim 1, wherein when the keyword is an industry name, the related words are a plurality of company names corresponding to the industry name. The industry correlation analysis system according to claim 1, wherein when the keyword is a company name, the processor executes the analysis program to further perform the following operations: According to the keyword, the processor is in the database A plurality of auxiliary related words are searched out, and according to one of the auxiliary related words, the processor searches for the related words in the database; wherein the auxiliary related words are a plurality of industry names corresponding to the company name . An industry association analysis method is applicable to an industry association analysis system. The industry association analysis system includes a user interface, a server, a memory, and a processor. The user interface is configured to input a key. And displaying an analysis result, the server is configured to run at least one database, the memory is configured to store an analysis program, the processor is connected to the user interface, the server and the memory, and the processor The method configured to execute the analysis program to perform the industry correlation analysis method includes: searching a plurality of related words in the database according to the keyword; searching the database for each preset time period and each A number of news articles related to related words, and based on the news, a word cloud is calculated and generated for each related word, where the word cloud includes multiple text features; the word features of the word clouds are compared to calculate and Each text feature has a probability of being associated with an overlap rate in the word clouds, and the text features of the word clouds are increased based on the overlap rate. Small sorting; and extracting several text features with a large overlap rate, and generating a correlation analysis diagram based on the extracted text features and the relationship between the extracted text features and the related words as the As a result of the analysis, a text feature with a large overlap rate is a text feature appearing in the word clouds at the same time, and the text feature appears in the word clouds multiple times, the text feature has a large overlap rate. The method for analyzing industrial relevance according to claim 6, further comprising: extracting a plurality of text features with a small overlap rate, and according to the text features being extracted and the text features being extracted and the associated words The relationship between the two results in another correlation analysis diagram as the analysis result. The method for analyzing industry correlation as described in claim 6, further comprising: extracting a plurality of text features with a large overlap rate and a plurality of text features with a small overlap rate, and according to the extracted text features and the The relationship between the extracted text features and the related words is used to generate another correlation analysis diagram as the analysis result. The method for analyzing industry relevance according to claim 6, wherein when the keyword is an industry name, the related words are a plurality of company names corresponding to the industry name. The method for analyzing industrial relevance according to claim 6, further comprising: according to the keyword, the processor searches for a plurality of auxiliary related words in the database, and according to one of the auxiliary related words, the processor The related words are searched in the database; among them, the auxiliary related words are a plurality of industry names corresponding to multiple company names.