JP4667823B2 - Table monitoring device, Web page monitoring system, computer program - Google Patents

Description

  The present invention relates to a table monitoring apparatus, a Web page monitoring system, and a computer program for realizing the table monitoring apparatus using a computer.

  In recent years, information services such as account information provision services for financial companies, stock information provision services, weather forecast information provision services, etc. have been provided by web pages established on websites on the Internet. The desired data can be acquired in a timely manner. When receiving such an information providing service, it is a heavy burden for the user himself / herself to always browse the web page and check whether the provided data has been updated. For this reason, it is desired to monitor specific data that the user pays attention to automatically detect that the specific data is updated and notify the user.

Conventionally, for example, a technique disclosed in Patent Document 1 is known as a technique for dividing a plurality of data included in a table in a Web page for each content. In this prior art, a table in a Web page is analyzed to generate cell position data indicating the position of a cell storing data and a cell vector expressing the characteristics of the cell, and the table type (table or layout) is set. Determine. If the table type is table, the table division direction (vertical division or horizontal division) is determined with reference to the cell position data and cell vector, the table is divided with reference to this division direction, and the segment Generate. On the other hand, when the table type is layout, each cell is clustered with reference to a cell vector, and a segment is generated with reference to cell position data and cell cluster information. Thereby, the table in the Web page is divided for each content.
JP 2000-339301 A

  However, in the above-described prior art, when the position of the cell storing the specific data fluctuates without being fixed in the table, it is difficult to uniquely identify the cell storing the specific data and is included in the table. It cannot be applied to the purpose of monitoring specific data.

  For example, in a stock price ranking table (see FIG. 9) that describes each brand information (code, market, name, transaction price, etc.) according to the ranking of the stock price increase rate, the ranking of the stock price increase rate of each stock that varies from time to time. Depending on the situation, the description position of each brand information varies. Therefore, in order to monitor the information of the specific brand that the user pays attention to, it is required to reliably identify the position of the specific brand in the table (in FIG. 9, the description line in the table). The brand information (code, name, transaction value, etc.) shown in FIG. 9 is for convenience.

  Further, when a plurality of brands are monitored in the stock price ranking table, and data of a plurality of brands to be monitored are collected and displayed, it is desirable that it is possible to easily determine which brand each display data is. For example, it is easier for the user to use “name” as information to be displayed for identifying display data than to use “code” in FIG.

  The present invention has been made in view of such circumstances, and the object of the present invention is to uniquely identify specific data in a table in which the position of a cell storing specific data varies, and for a user. It is an object of the present invention to provide a table monitoring apparatus capable of selecting identification information that can be easily understood.

  Another object of the present invention is to automatically monitor the update of specific data that is noticed by the user by using the table monitoring apparatus of the present invention to monitor specific data included in a table in a Web page. Another object of the present invention is to provide a Web page monitoring system that can detect the target and collect the specific data and display it easily for the user.

  Another object of the present invention is to provide a computer program for realizing the table monitoring apparatus of the present invention using a computer.

  In order to solve the above problems, a table monitoring apparatus according to the present invention is a table monitoring apparatus that monitors a table that forms a matrix with a plurality of cells each storing data, and includes an example table, an analysis target table, And an input means for inputting designated data for specifying attention data in the example table, and an identifier row or an identifier column based on the similarity between the data of the example table and the analysis target table from the example table. Selecting and extracting an object including data of the identifier row or identifier column from the example table, an identifier / object extracting means for defining the identifier row or identifier column data as an identifier of the object, and the example Extract a positive example object including the attention data from the table and analyze Object identifying means for extracting an object of interest including an identifier that matches the identifier of the positive example object from a table, and identifying the data at the same position as the data of attention in the positive example object as the target data from the target object When a plurality of identifier row candidates or identifier column candidates are found in the identifier extraction process performed by the attention data identification means and the identifier / object extraction means, a keyword is selected from the text included in the plurality of candidate cells. The search means for transmitting the keyword to the Web search site and receiving this response and the search result number for each keyword are obtained based on the response received by the search means, and the search result number for each keyword is compared respectively. Based on this comparison result, the optimum identifier row candidate or identifier column Select, characterized by comprising an identifier selection means for notifying an identifier line candidate or the identifier string candidates of the selection result to the identifier object extracting means.

  In the table monitoring apparatus according to the present invention, the search means excludes, from among the plurality of candidates, a candidate including a cell whose length of a character string of text in a cell exceeds a predetermined number of characters from a Web search target. It is characterized by.

  In the table monitoring apparatus according to the present invention, the search means may exclude, from the plurality of candidates, a candidate including a cell whose text in a cell is only a numerical value and exceeds a predetermined number of digits, from a Web search target. Features.

  In the table monitoring apparatus according to the present invention, the identifier selection unit calculates an average value of the number of search results for each candidate, compares the average value of the number of search results for each candidate, and calculates the minimum average value. The candidate is selected.

  In the table monitoring apparatus according to the present invention, the identifier selection unit excludes the maximum value and the minimum value from the search result number in the average value calculation process of the search result number, and averages the number of remaining search results. It is characterized by calculating a value.

  The web page monitoring system according to the present invention is a web page monitoring system that monitors a web page having a table that forms a matrix with a plurality of cells each storing data, the web page for illustration and the analysis target Web page acquisition means for acquiring a Web page via a communication network, attention data designation input means for designating attention data in the example Web page, a table in the example Web page, and the analysis target Web page The table monitoring device according to any one of claims 1 to 5 is identified by the table monitoring device and the table monitoring device. Storage means for storing received attention data, and the storage means From the memorized attention data and the attention data obtained by the table monitoring device, determination means for determining whether or not attention data in the analysis target Web page is updated, and the determination result of the determination means to the user And a display data creating means for creating display data for displaying the attention data identified by the table monitoring device and the identifier used for the identification in association with each other. And

A computer program according to the present invention is a computer program for monitoring a table that forms a matrix with a plurality of cells each storing data, and includes an example table, an analysis target table, and data of interest in the example table The identifier row or identifier column based on the similarity between the data of the example table and the analysis target table is selected from the example table, and the identifier row or identifier column of the identifier row or identifier column is selected. An object including data is extracted from the example table, and an identifier / object extraction function for defining the data of the identifier row or identifier column as the identifier of the object, and a positive example object including the attention data from the example table Extracted from the analysis target table A function of extracting an attention object including an identifier that matches an identifier of a positive example object; a function of identifying data at the same position as the position of attention data in the positive example object as attention data from the attention object; When a plurality of identifier row candidates or identifier column candidates are found in the identifier extraction process performed by the identifier / object extraction function , a keyword is selected from text included in the plurality of candidate cells, and the keyword is sent to the Web search site. The function of transmitting and receiving this response and the number of search results for each keyword are obtained based on the received response, the number of search results for each keyword is compared, and an optimal identifier row candidate or Select an identifier column candidate, and select an identifier row candidate or an identifier column candidate as a result of this selection. A function of notifying the serial identifier object extracting function, characterized in that to realize the computer.
As a result, the table monitoring apparatus described above can be realized using a computer.

  According to the present invention, attention is focused on the similarity in data between tables including data that changes or moves from moment to moment, and attention data in the example table is identified by identifying the attention data after identifying each object. It is possible to identify the data of interest in the analysis target table simply by presenting an example of the data. Thereby, it is possible to automatically detect the presence of update by monitoring specific data. Furthermore, an identifier selection process using Web search can select a meaningful identifier that can be easily understood by the user so that the identifier can be used as a title of display data.

  In addition, by using the table monitoring apparatus of the present invention to monitor specific data included in a table in a Web page, it is possible to automatically detect that there is an update of specific data that the user is interested in and Data can be collected for easy-to-understand display for the user.

First, preconditions for the table handled by the present invention will be described.
Each table has a plurality of cells for storing data, and a plurality of cells form a matrix.
The table has several objects of the same type. For example, it deals with a table in which rankings such as stock trading volume and weekly weather forecasts are described. An object is composed of a plurality of cells.
The same kind of data is arranged in the column direction and data related to the row direction are arranged, and one or more objects are arranged in one row. Alternatively, the same kind of data is arranged in the row direction and data related to the column direction are arranged, and one or more objects are arranged in one column.
Generation, deletion, and movement within the table are performed in units of objects.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the present embodiment, an apparatus that monitors specific data included in a table in a Web page will be described as an example.
FIG. 1 is a block diagram showing a configuration of a table monitoring apparatus 1 according to an embodiment of the present invention. In FIG. 1, the table monitoring apparatus 1 includes an input unit 11, an identifier / object extraction unit 12, an object identification unit 13, an attention data identification unit 14, an output unit 15, an exemplary web page update unit 16, a search unit 17, and an identifier selection. The unit 18 is provided.

  The input unit 11 inputs web page data. Web page data is, for example, HTML (HyperText Markup Language) data. Data is stored in a table format in the Web page. There are two types of Web pages that are input, for example and for analysis. The exemplary web page is a page used when designating data (attention data) to which the user pays attention. Further, the input unit 11 inputs designation data for specifying the attention data. The designation data may be the attention data. The analysis target web page is a page that is a target for actually identifying the attention data. The example web page and the analysis target web page are pages having the same URL (Uniform Resource Locator).

  The identifier / object extraction unit 12 compares the same rows with each other in the tables of the example Web page and the analysis target Web page, and also compares the same columns with each other, and completely matches the same row or the same column. From the table of the example Web page, select one of the rows or columns for which there are no two or more texts to be displayed. If a row is selected here, the row is set as an identifier row, and if a column is selected, the column is set as an identifier column. Then, from the cells in the table of the exemplary web page, all cells or a part of the cells in the column direction for the identifier row and in the row direction for the identifier column are cut out as objects. The cell at the intersecting position is used as the identifier of the object.

  The search unit 17 and the identifier selection unit 18 are blocks that perform an identifier selection process using Web search. The identifier selection process using the Web search is performed when an identifier extraction process performed by the identifier / object extraction unit 12 finds a plurality of identifier row / identifier column candidates, and an identifier that can be easily understood by the user from the plurality of candidates. This is for selecting a candidate including.

  The search unit 17 transmits a query (keyword to be searched) to the Web search site and receives the response. The keyword is selected from text included in a plurality of identifier row / identifier column candidate cells discovered by the identifier / object extraction unit 12.

  The identifier selection unit 18 obtains the number of search results (number of search hits) for each keyword based on the response received by the search unit 17. Then, the number of search results for each keyword is compared, and an optimum identifier row / identifier column candidate is selected based on the comparison result. The identifier row / identifier column candidates of this selection result are notified to the identifier / object extraction unit 12.

  The object identification unit 13 extracts an object (positive example object) including attention data from the exemplary Web page. Then, an object including an identifier that matches the identifier of the positive example object is cut out from the analysis target Web page, and this object is set as an object (attention object) including attention data. If there is no object that matches the positive object in the analysis target Web page, an error is assumed and identification is impossible.

The attention data identification unit 14 identifies, as attention data, data that is at the same position as the position of attention data in the positive example object from the attention object.
The output unit 15 outputs the attention data in the analysis target Web page cut out by the attention data identification unit 14.

  In order to correctly extract the object identifier, the example Web page update unit 16 performs the above-described processing (identifier / object extraction) at a constant multiple of the information update period (for example, one day or one hour) of the Web page. , Identification of object, identification of attention data). If the data of interest can be identified, the analysis target Web page is updated as a new example Web page. Note that the number of constant multiplications is changed according to the communication cost and load.

  As described above, the table monitoring apparatus 1 extracts a collection (object) of related data using the characteristics of the data structure of the table in the Web page, identifies the object including the attention data, and then extracts the attention data. Identify. As a result, even if the value of the attention data specified by the user changes or the position of the attention data moves in the table, it is only necessary to present an example of the attention data in the example Web page. The attention data can be identified from the analysis target Web page, and the attention data can be automatically and continuously acquired from the Web page.

  The table monitoring apparatus according to the present embodiment may be realized by dedicated hardware, or may be configured by a general-purpose computer system such as a personal computer, and each function of the apparatus shown in FIG. You may implement | achieve the function by running the program for implement | achieving.

In addition, an input device, a display device, and the like (none of which are shown) are connected to the table monitoring device as peripheral devices. Here, the input device refers to an input device such as a keyboard and a mouse. The display device refers to a CRT (Cathode Ray Tube), a liquid crystal display device or the like.
The peripheral device may be connected directly to the table monitoring device, or may be connected via a communication line.

Next, the detailed operation of the table monitoring apparatus 1 shown in FIG. 1 will be described. 2 to 7 are process flow diagrams of the table monitoring apparatus 1 shown in FIG.
First, in FIG. 2, the user designates attention data while viewing the example Web page (Web1) with the user terminal (step S101). For example, the cell number of the cell having the attention data is input by keyboard operation, or the mouse is clicked by operating the mouse or the like. This designation data is input to the table monitoring apparatus 1. Next, the table monitoring apparatus 1 takes out the table T1 including the designated attention data from the example web page (Web1) based on the inputted designated data (step S102). Next, the table monitoring apparatus 1 acquires an analysis target Web page (Web2) (step S103). Next, the same table T2 as the table T1 is taken out from the analysis target Web page (Web2) (step S104). Next, the identifier column or identifier row of the object is extracted from the tables T1 and T2 (step S105).

Here, the identifier extraction processing in step S105 will be described with reference to FIGS.
First, it is determined whether the identifier is in the row direction or the column direction.
In FIG. 4, the “(row, column) = (i, j)”-th cell of the table T1 is compared with the cell of the table T2 (step S201), and “(row, column) = (i, j, j) The number of cells in the table T2 that completely match the character string of the "th cell" is calculated for all (i, j) under the following conditions A1 and A2, and two tables (row direction table, column) (Direction table) is generated (step S202).
Condition A1: Count the number of character strings in the (i, j) th cell of the table T1 included in the i-th row of the table T2. The count result of condition A1 is stored in the (i, j) th cell in the row direction table. As a result, a row direction table having the same number of matrices as the table T1 is generated.
Condition A2: The number of character strings in the (i, j) th cell of the table T1 included in the jth column of the table T2 is counted. The count result of condition A2 is stored in the (i, j) th cell in the column direction table. As a result, a row direction table having the same number of matrices as the table T1 is generated.

  Next, either the row direction table or the column direction table is selected by the procedure of steps S203 to S209. In step S203, it is determined whether there is a table including two or more numerical values. If the result of this determination is “YES”, the table containing the numerical value of 2 or more is selected (step S204). On the other hand, if the determination result in step S203 is “NO”, the sum of the numerical values in the table is calculated for each table, and these total values are compared to determine whether the total values are different (step S205). If the result of this determination is “YES”, the table with the larger total value is selected (step S206).

On the other hand, if the determination result in step S205 is “NO”, the table is selected according to the following conditions B1 to B4.
Condition B1: When there is a cell containing the same character string in the same row of the table T1, the row direction table is selected.
Condition B2: If there is a cell containing the same character string in the same column of the table T1, the column direction table is selected.
Condition B3: If there is a cell containing the same character string in the same row of the table T2, the row direction table is selected.
Condition B4: If there is a cell containing the same character string in the same column of the table T2, the column direction table is selected.
It is determined whether any of the above conditions B1 to B4 is met (step S207). If the conditions are met, the table is selected under the matched conditions (step S208).

  On the other hand, if none of the above conditions B1 to B4 is met (NO in step S207), the degree of similarity between the cells in the table is calculated using the table T1 or T2, and the direction in which the degree of similarity is high is the row direction. Or a column direction is selected, and a table in a direction equal to the direction having a high similarity is selected (step S209). The method for calculating the similarity will be described later.

  When a row direction table or a column direction table is selected by the above-described procedure, an identifier row or an identifier column is next obtained by the process of FIG. 5 based on the selected table. That is, if the row direction table is selected, the identifier row is obtained. On the other hand, if the column direction table is selected, an identifier column is obtained. In the following description, the case of obtaining an identifier row from a row direction table is described as an example, but the same applies to the case of obtaining an identifier column from a column direction table, where rows are columns, top is left, bottom is right In addition, the right may be read down and the left up. The description in parentheses in FIG. 5 corresponds to the case where the identifier column is obtained from the column direction table.

  In FIG. 5, first, a row including two or more values in the row direction table is excluded from the identifier row candidates (step S210). As a result, when two or more identifier row candidates still remain (YES in step S211), when there are a plurality of cells containing the same character string in the same row of the table T1, the row is further excluded from the identifier row candidates. (Step S212). As a result, when two or more identifier row candidates still remain (YES in step S213), when there are a plurality of cells including the same character string in the same row of the table T2, the row is further excluded from the identifier row candidates. (Step S214).

  As a result, when two or more identifier row candidates still remain (YES in step S215), a multiple identifier handling process is executed (step S216). As a result, if two or more identifier row candidates still remain (YES in step S217), a row including the point of interest (attention data designated by the user) is further excluded from the identifier row candidates (step S218). . As a result, when two or more identifier row candidates still remain (YES in step S219), the multiple identifier handling process is executed again (step S220). As a result, if two or more identifier row candidates still remain (YES in step S221), an identifier selection process using Web search is executed (step S222). Thereafter, the process proceeds to step S242. The multiple identifier handling process (S216, S220) and the identifier selection process (S222, S241) using Web search will be described later.

On the other hand, if there are less than two identifier row candidates in step S211 (NO in step S211), it is determined whether there are 0 identifier row candidates (step S230). If YES in S230, the process is terminated with no identifier (step S231). If there is one identifier row candidate (NO in step S230), the process proceeds to step S242.
Further, when there are less than two identifier row candidates in any of the above steps S213, S215, and S219 (NO in any of steps S213, S215, and S219), it is determined whether there are 0 identifier row candidates ( Step S240). If the number of identifier row candidates is zero as a result of this determination (YES in step S240), the previous candidate exclusion process is returned to the canceled state, and the identifier selection process using Web search is executed ( Step S241). On the other hand, if NO in step S240, the process proceeds to step S242.
If there are less than two identifier row candidates in any of the above steps S217 and S221 (NO in any of steps S219 and S221), the process proceeds to step S242.
Next, in step S242, an identifier row selected as a result of the above processing is set.

  The identifier row or the identifier column is extracted by the identifier extraction process in step S105 described above.

  Returning to FIG. In step S106, it is determined whether an identifier row or an identifier column has been set. If it has not been set (NO in step S106), an error is output and the process ends. On the other hand, when an identifier row or an identifier column is set (YES in step S106), the process proceeds to step S107.

In the following description, the case where an identifier column is extracted by the above-described identifier extraction processing will be described as an example. However, the same applies to the case where an identifier row is extracted, and the column may be read as a row.
In step S107, all or some of the cells in the row direction are extracted from the table T1 as objects, and the text of the cell that intersects the identifier column for each object is assigned as the identifier (object identifier) of the object.

  Next, in step S110 of FIG. 3, an object in the row direction including the same text as the object identifier of the object in the table T1 including the target point is searched from the table T2. If no corresponding object is found as a result of this search (NO in step S111), an error is output and the process ends. On the other hand, when a corresponding row-direction object is found, a cell in the same position (in the same column) as the cell of the target point in the object of the table T1 among the cells in the object is selected as the target point of the table T2. To the user terminal (step S112).

  Next, with reference to FIG. 6, the above-described multiple identifier handling process (steps S216 and S220 in FIG. 5) will be described. In the following description, a case where there are a plurality of identifier column candidates will be described as an example. However, the same applies when there are a plurality of identifier row candidates. Should be read side by side. The description in parentheses in FIG. 6 corresponds to the case where there are a plurality of identifier row candidates.

In FIG. 6, first, each similarity is calculated for all combinations of identifier strings (step S301). This similarity calculation method will be described later. Next, the similarity is compared with a predetermined threshold, and if at least one similarity is equal to or greater than the threshold (YES in step S302), the identifier strings are similar and the object can be divided into a plurality of objects. Therefore, the process proceeds to step S303, and further processing is executed.
On the other hand, if all the similarities are less than the threshold (NO in step S302), the identifier strings are not similar to each other, and the process is terminated as it is.

  In step S303, the variable i is set to an initial value “1”. Next, the similarity between i-numbered right columns is calculated for each similar identifier column (step S304), the similarity is compared with a threshold value, and if it is less than the threshold value, the process is terminated as it is (step S305). NO). On the other hand, if it is equal to or greater than the threshold value, 1 is added to the variable i (step S306). If the i-th column is an identifier column or there is no i-th right column (YES in step S307), the process proceeds to step S308; otherwise (NO in step S307), the process returns to step S304. As a result, the similarity between all similar identifier columns and all columns sandwiched between adjacent identifier columns is calculated. Next, in step S308, it is determined whether or not all the similarities are equal to or greater than a threshold. If the total similarity is equal to or greater than the threshold, the object is set to be divided by two or more identifier strings. Next, the table T1 is divided between the (i−1) -th and the i-th to the right from the identifier column, and the two divided tables are connected in the vertical direction sequentially from the left to generate a new table T1 (step S309). ). Similarly, the table T2 is connected in the vertical direction after being divided, and a new table T2 is generated (step S309).

Next, the similarity calculation process described above will be described with reference to FIG.
In FIG. 7, first, for one of the comparison targets, the character string in the cell is extracted, and an initial value “1” is set to the variable i (step S401). Next, the i-th character in the extracted character string is extracted. If this character is a numerical value, it is “N”, if it is an img tag, it is “I”, if it is an a tag, it is “A”, and if it is other text. Replace with “T” (steps S403 to S409). For example, the character string “March 23 (Tuesday)” is converted to the replacement string “NTNNTTTTT”. Next, 1 is added to the variable i (step S410). If the i-th character is in the extracted character string, the process returns to step S402 (YES in step S411), and if not, the other comparison target is also in step S401. To S411 (NO in step S411, step S412). Next, for each comparison target replacement sequence, the degree of similarity is calculated from the combination of symbols and the length of the symbol sequence using DP matching.

  Next, an identifier selection process (steps S222 and S241 in FIG. 5) using the above-described Web search will be described with reference to FIG. In the following description, a case where there are a plurality of identifier row candidates will be described as an example, but the same applies when there are a plurality of identifier column candidates, and the rows can be read as columns and the top as left. That's fine. The description in parentheses in FIG. 8 corresponds to the case where there are a plurality of identifier string candidates.

  In FIG. 8, first, of the plurality of identifier row candidates, a row including a cell in which the length of the character string of the text in the cell exceeds L characters is excluded from the identifier row candidates (step S501). L is a value of the number of characters to be limited (for example, 10). By this candidate exclusion process, a row of a cell storing text inappropriate for an identifier such as an explanatory note is excluded from the identifier row candidates.

  As a result, when two or more identifier row candidates still remain (YES in step S502), a row including a cell whose text in the cell is only numeric and exceeds M digits is excluded from the identifier row candidates ( Step S503). M is a value of the number of digits to be limited (for example, 5). By this candidate exclusion process, a row of a cell that stores a numerical value with a number of digits inappropriate for an identifier, which is difficult for humans to discriminate, is excluded from identifier row candidates.

  As a result, when two or more identifier row candidates still remain (YES in step S504), the process proceeds to step S508. On the other hand, if there are less than two identifier row candidates (NO in step S502 or S504) by the candidate exclusion process (S501, S503), it is determined whether there are 0 identifier row candidates (step S505). As a result of this determination, if there are no identifier line candidates (YES in step S505), the process returns to the state in which the immediately preceding candidate exclusion process is canceled, and the process proceeds to step S508. On the other hand, if “NO” in the step S505, that is, if the identifier row candidates are narrowed down to one, the process proceeds to a step S507 to select the identifier row candidate (step S507).

  In step S508, the number of cells included in one row is counted for each remaining identifier row candidate (step S508). Next, the count value n is compared with a threshold value N (for example, N = 10) to determine whether the count value n is less than the threshold value N (step S509). According to this result, the following case classification (cases A and B) is performed, and a Web search is performed using text in a cell included in the identifier row candidate as a keyword.

Case A;
When the count value n is less than the threshold value N (YES in step S509), all the cells included in the identifier row candidate are set as keyword selection targets, and the text included in one cell of the keyword selection target is selected as one keyword. As a result, as many keywords as the number of cells included in the identifier row candidate are selected. Then, the keywords for the selected number of cells are transmitted to the Web search site, and the search results (for the number of cells) for each keyword are received (step S510). This keyword search is performed for each identifier row candidate, and a search result for each identifier row candidate is obtained.

Case B;
When the count value n is greater than or equal to the threshold value N (NO in step S509), among the cells included in the identifier row candidate, N cells counted from the top are set as keyword selection targets and included in one cell of the keyword selection target. Selected text as one keyword. As a result, N keywords are selected. Then, the selected N keywords are transmitted to the Web search site, and the search results (N) for each keyword are received (step S511). This keyword search is performed for each identifier row candidate, and a search result for each identifier row candidate is obtained.

  Next, an average value of the number of search results is calculated for each identifier row candidate (step S512). In the average value calculation process for the number of search results, the maximum and minimum values are excluded from the number of search results for each keyword (number of search hits) for one identifier row candidate, and the average value of the remaining search results is calculated. To do. By excluding the maximum value and the minimum value, it is possible to obtain an appropriate average value excluding the number of unique search results.

  Next, the average value of the number of search results for each identifier row candidate is compared, and the identifier row candidate having the smallest average value is selected (step S513). Here, the reason why the identifier row candidate having the smallest average value is selected is that the keyword is used because the use of the keyword is more limited when the number of search results (number of search hits) is smaller. This is because the number of meanings is small. That is, as the identifier has a smaller number of meanings, it is easier for the user to associate the meaning of the keyword and to determine it. Note that the numerical value has a different meaning for each scene used, and it is difficult to determine what the meaning is. On the other hand, proper nouns such as company names are easy to distinguish because their meanings are almost limited.

  For example, in the stock price ranking table of FIG. 9, in the identifier extraction process of FIGS. 4 and 5, the “code” row and the “name” row are identifier row candidates. Next, a web search is performed using the text in the cell of each identifier row candidate as a keyword by the identifier selection process using the web search of FIG. 8 described above. As a result, the number of search results (number of search hits) is smaller when “name” is a keyword than when “code” is a keyword. As a result, the “name” line is selected as the identifier line, and “name” becomes the identifier. As a result, it is possible to select an identifier that is easily understood by the user, that is, the name of the brand.

  As described above, according to the present embodiment, specific data can be uniquely identified in a table in which the position of a cell storing specific data varies. Furthermore, identification information that can be easily understood by the user can be selected by identifier selection processing using Web search.

  FIG. 10 is a block diagram showing the configuration of the web page monitoring system 20 according to one embodiment of the present invention. 10, the Web page monitoring system 20 includes the table monitoring apparatus 1, the Web page acquisition unit 21, the attention data designation input unit 22, the attention data storage unit 23, the comparison unit 24, the notification unit 25, and the display data generation unit illustrated in FIG. 1. 26.

The Web page acquisition unit 21 acquires an exemplary Web page and an analysis target Web page via a communication network using a URL specified by the user. The exemplary web page is output to the attention data designation input unit 22 and the table monitoring apparatus 1. The analysis target web page is output to the table monitoring apparatus 1.
The attention data designation input unit 22 has a function of designating attention data by a user operation from the exemplary Web page acquired by the Web page acquisition unit 21, and outputs data specifying the attention data to the table monitoring device 1.

The attention data storage unit 23 stores attention data in the table of the analysis target Web page identified by the table monitoring apparatus 1.
The comparison unit 24 compares the attention data in the table of the analysis target Web page obtained this time by the table monitoring device 1 with the attention data previously stored in the attention data storage unit 23, and the comparison result is notified to the notification unit 25. Output to. Further, when there is a change in the attention data, new attention data is output to the display data creation unit 26.

  The notification unit 25 notifies the user of whether or not attention data in the analysis target Web page is updated based on the comparison result of the comparison unit 24. For example, when there is an update, notification is made by e-mail. At this time, the updated latest attention data may be described in the e-mail. Thereby, it is possible to monitor the specific data included in the table in the Web page using the table monitoring device 1 and automatically detect the update of the specific data that the user pays attention to.

  The display data creation unit 26 creates display data using the attention data received from the comparison unit 24. In this display data creation process, the display data is configured to display the corresponding identifier in association with the data that the user pays attention to. This identifier is used by the table monitoring apparatus 1 to identify the data of interest, and is easily understood by the user.

  For example, if the cell of interest specified by the user in the stock price ranking table of FIG. 9 is a “volume” cell, the data of interest is the value of the volume. Then, a set of identifier and volume value is output from the table monitoring apparatus 1. The display data creation unit 26 acquires this reception date and time (data acquisition date and time) when receiving attention data from the comparison unit 24. Then, display data for displaying the identifier, the data acquisition date and time, and the volume value in association with each other is created. FIG. 11 is a display example of attention data. In the example of FIG. 11, a brand name (for example, XX Industrial Co., Ltd.) is displayed as an identifier in <> on the screen 100 of the terminal that displays the display data. Then, in association with the identifier (name of brand), the value of the volume and the acquisition date / time are displayed in the order of the acquisition date / time and the value of the volume, such as “○ month × day 11:30, 12300”.

  As described above, in the present embodiment, the table monitoring device 1 selects an identifier (in FIG. 11, the name of the brand) that can be easily understood by the user. For example, in the example of FIG. 11, the identifier (name of brand) is displayed in association with the volume value, so that the user can easily determine which brand the volume value displayed is for. Can be determined.

  As described above, according to the present embodiment, specific data that is noticed by the user is updated by monitoring specific data included in the table in the Web page using the table monitoring apparatus 1 of FIG. Can be detected automatically. Furthermore, the specific data can be collected and displayed in an easy-to-understand manner for the user.

Note that the table monitoring is performed by recording a program for realizing each step shown in FIGS. 2 to 8 on a computer-readable recording medium, causing the computer system to read and execute the program recorded on the recording medium. Processing may be performed. Here, the “computer system” may include an OS and hardware such as peripheral devices.
Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used.
The “computer-readable recording medium” means a flexible disk, a magneto-optical disk, a ROM, a writable nonvolatile memory such as a flash memory, a portable medium such as a CD-ROM, a hard disk built in a computer system, etc. This is a storage device.

Further, the “computer-readable recording medium” means a volatile memory (for example, DRAM (Dynamic DRAM) in a computer system that becomes a server or a client when a program is transmitted through a network such as the Internet or a communication line such as a telephone line. Random Access Memory)), etc., which hold programs for a certain period of time.
The program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line.
The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes design changes and the like within a scope not departing from the gist of the present invention.

It is a block diagram which shows the structure of the table monitoring apparatus 1 which concerns on one Embodiment of this invention. It is a processing flow figure of the table monitoring apparatus 1 shown in FIG. It is a processing flow figure of the table monitoring apparatus 1 shown in FIG. It is a processing flow figure of the table monitoring apparatus 1 shown in FIG. It is a processing flow figure of the table monitoring apparatus 1 shown in FIG. It is a processing flow figure of the table monitoring apparatus 1 shown in FIG. It is a processing flow figure of the table monitoring apparatus 1 shown in FIG. It is a processing flow figure of the table monitoring apparatus 1 shown in FIG. It is a figure which shows the specific example of the table of monitoring object. It is a block diagram which shows the structure of the web page monitoring system 20 which concerns on one Embodiment of this invention. It is an example of a display of attention data by display data which Web page monitoring system 20 shown in Drawing 10 creates.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Table monitoring apparatus, 11 ... Input part, 12 ... Identifier / object extraction part, 13 ... Object identification part, 14 ... Attention data identification part, 15 ... Output part, 16 ... Example Web page update part, 17 ... Search part , 18 ... identifier selection section, 20 ... Web page monitoring system, 21 ... Web page acquisition section, 22 ... attention data designation input section, 23 ... attention data storage section, 24 ... comparison section, 25 ... notification section, 26 ... display data Creation department.

Claims (7)

A table monitoring device for monitoring a table constituting a matrix by a plurality of cells each storing data,
An input unit for inputting specification data for specifying the table of interest, the analysis target table, and the attention data in the table of illustration;
An identifier row or identifier column based on the similarity between the data of the example table and the analysis target table is selected from the example table, and an object including data of the identifier row or identifier column is selected from the example table. And an identifier / object extraction means for defining the identifier row or identifier column data as the identifier of the object,
An object identification means for extracting a positive example object including the attention data from the example table and extracting an attention object including an identifier that matches the identifier of the positive example object from the analysis target table;
Attention data identification means for identifying, as attention data, data at the same position as the attention data position in the positive example object from the attention object;
When a plurality of identifier row candidates or identifier column candidates are found in the identifier extraction process performed by the identifier / object extraction means, a keyword is selected from texts included in the plurality of candidate cells, and the keyword is selected as a Web search site. Search means to send to and receive this response;
Based on the response received by the search means, obtain the number of search results for each keyword, compare the number of search results for each keyword, select the optimal identifier row candidate or identifier column candidate based on the comparison result, Identifier selection means for notifying the identifier / object extraction means of an identifier row candidate or an identifier column candidate of the selection result;
A table monitoring apparatus comprising:   2. The search unit according to claim 1, wherein the search unit excludes, from the Web search target, a candidate including a cell in which a length of a character string of a text in a cell exceeds a predetermined number of characters among the plurality of candidates. Table monitoring device.   2. The table according to claim 1, wherein the search unit excludes, from the Web search target, candidates including a cell whose text in a cell is only a numerical value and exceeds a predetermined number of digits, among the plurality of candidates. Monitoring device.   The identifier selecting means calculates an average value of the number of search results for each candidate, compares the average value of the number of search results for each candidate, and selects the candidate having the minimum average value. The table monitoring apparatus according to claim 1.   The identifier selecting means, in the average value calculation process of the search result number, excludes a maximum value and a minimum value from the search result number and calculates an average value of the remaining search result number. Item 5. The table monitoring device according to Item 4. A web page monitoring system for monitoring a web page having a table that forms a matrix with a plurality of cells each storing data,
Web page acquisition means for acquiring the Web page for analysis and the Web page to be analyzed via a communication network;
Attention data designation input means for designating attention data in the exemplary web page;
The attention data in the table of the analysis target Web page is identified from the table in the example Web page, the table in the analysis target Web page, and the specified data of the attention data. The table monitoring device according to the section;
Storage means for storing attention data identified by the table monitoring device;
Determination means for determining whether or not the attention data in the analysis target Web page is updated from the attention data stored in the storage means and the attention data obtained by the table monitoring device;
Notification means for notifying the user of the determination result of the determination means;
Display data creating means for creating display data for displaying the attention data identified by the table monitoring device and the identifier used for the identification in association with each other;
A web page monitoring system comprising: A computer program for monitoring a table that forms a matrix with a plurality of cells each storing data,
An example table, an analysis target table, and a function of inputting designation data for specifying attention data in the example table;
An identifier row or identifier column based on the similarity between the data of the example table and the analysis target table is selected from the example table, and an object including data of the identifier row or identifier column is selected from the example table. And an identifier / object extraction function for extracting the data of the identifier row or identifier column as the identifier of the object ,
A function of extracting a positive example object including the attention data from the example table, and extracting a target object including an identifier that matches the identifier of the positive example object from the analysis target table;
A function for identifying, as attention data, data at the same position as the attention data position in the positive example object from the attention object;
When a plurality of identifier row candidates or identifier column candidates are found in the identifier extraction processing performed by the identifier / object extraction function , a keyword is selected from texts included in the plurality of candidate cells, and the keyword is selected as a Web search site. To send and receive this response,
Based on the received response, the number of search results for each keyword is obtained, the number of search results for each keyword is compared, and an optimum identifier row candidate or identifier column candidate is selected based on the comparison result. A function for notifying an identifier row candidate or an identifier column candidate to the identifier / object extraction function;
A computer program for causing a computer to realize the above.

Images