Similarity calculation device, similarity calculation method, and similarity calculation program

Abstract

An extraction unit (15c) extracts a part that represents ID from each of two URLs which are to be processed and which are included in an operation log. A determination unit (15d) determines whether the part that represents the ID is a temporarily generated part, by using statistical information in the operation log in a prescribed period. A calculation unit (15e) calculates the similarity between the two URLs to be processed, excluding the part that represents the ID, when the part that represents the ID is a temporarily generated part.

Description

Technical Field

[0001] The present invention relates to a similarity calculation device, a similarity calculation method, and a similarity calculation program.

Background Art

[0002] In the automation of operations on a PC and the analysis of operations on a PC, such as RPA (Robotic Process Automation), it is necessary to determine the identity of web pages. In this case, in addition to comprehensively using various information such as window titles and page contents, the URL (Uniform Resource Locator) becomes particularly important information.

[0003] In recent years, web sites have become highly functional and their mechanisms have become complex. Since the URLs representing individual web pages that make up a web site contain various information, there are cases where the URL changes even for the same web page. Regarding the comparison of URLs necessary for determining the identity of web pages, it cannot be determined only by whether the URLs exactly match or not, and it is necessary to make a determination considering the similarity of the URLs. Conventionally, the similarity of URLs has been evaluated using, for example, the ratio of matching when comparing the URL strings from the front, the edit distance (Levenshtein distance), etc. (see Non-Patent Document 1).

Prior Art Documents

Non-Patent Documents

[0004]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the prior art, the proximity of the actual URLs may not be fully reflected in the similarity. For example, when a system-specific ID is included in a URL, it is difficult to correctly evaluate the similarity of web pages.

[0006] The present invention has been made in view of the above, and an object thereof is to enable highly accurate evaluation of the similarity of URLs for use in determining the identity of web pages.

Means for Solving the Problems

[0007] In order to solve the above-described problems and achieve the object, a similarity calculation device according to the present invention includes an extraction unit that extracts a part representing an ID from each of two URLs to be processed included in an operation log, a determination unit that determines whether or not the part representing the ID is a part that is temporarily generated using statistical information in the operation log for a predetermined period, and a calculation unit that excludes the part representing the ID when the part representing the ID is a part that is temporarily generated and calculates the similarity of the two URLs to be processed.

Effects of the Invention

[0008] According to the present invention, it is possible to highly accurately evaluate the identity of a web page in consideration of the similarity of URLs.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

[0010] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited by this embodiment. In the description of the drawings, the same parts are denoted by the same reference numerals.

[0011] [Outline of Similarity Calculation Device] FIGS. 1 and 2 are diagrams for explaining the outline of the similarity calculation device. The similarity calculation device calculates the similarity of the URL string in order to determine the identity of Web pages. For example, as illustrated in FIG. 1, any two URLs included in the operation log are compared to calculate the similarity. At that time, the similarity calculation device accurately evaluates the similarity of the URLs by statistically using the past operation log.

[0012] Here, conventionally, when comparing URLs, the similarity is evaluated using the matching ratio from the beginning of the URL string and the edit distance (Levenshtein distance) representing the number of procedures required for replacement. However, when a system-specific ID is included in the URL, the semantic proximity of the actual URL cannot be evaluated.

[0013] For example, as illustrated in FIG. 2(a), when some ID is included in the URL, different Web pages may be determined to be close, that is, have a high similarity. Also, when a temporary ID that is temporarily generated and discarded is included in the URL, the same Web page may be determined to be far, that is, have a low similarity.

[0014] Therefore, as illustrated in FIG. 2(b), the similarity calculation device according to the present embodiment decomposes the URL into elements according to the http / https scheme syntax and detects the ID part in the URL in consideration of the characteristics of the ID. Then, for the detected ID, the similarity calculation device determines that it is a temporary ID such as a session ID when, for example, it has not been reused across days, considering the appearance frequency in past logs.

[0015] Here, many of the temporary IDs are IDs for managing sessions and are often not related to the identity of the web page represented by the URL. On the other hand, other persistent IDs are often strongly related to the web page represented by the URL. Therefore, the similarity calculation device assumes that the temporary ID has no influence on the similarity of the two URLs to be evaluated, and increases the weight on the similarity for the other persistent IDs to calculate the similarity of the two URLs.

[0016] In this way, the similarity calculation device can accurately determine the similarity of the URLs required for determining the identity of the web page. Note that the processing target of the similarity calculation device is not limited to URLs and may be URIs (Uniform Resource Identifiers) or URNs (Uniform Resource Names).

[0017] [Configuration of Similarity Calculation Device] FIG. 3 is a schematic diagram illustrating the schematic configuration of the similarity calculation device. As illustrated in FIG. 3, the similarity calculation device 10 according to the present embodiment is realized by a general-purpose computer such as a personal computer and includes an input unit 11, an output unit 12, a communication control unit 13, a storage unit 14, and a control unit 15.

[0018] The input unit 11 is realized using an input device such as a keyboard or a mouse, and inputs various instruction information such as a processing start to the control unit 15 in response to an input operation by an operator. The output unit 12 is realized by a display device such as a liquid crystal display, a printing device such as a printer, and the like. For example, the result of the similarity calculation process described later is displayed on the output unit 12.

[0019] The communication control unit 13 is implemented by a NIC (Network Interface Card) or the like, and controls communication between an external device and the control unit 15 via a telecommunications line such as a LAN (Local Area Network) or the Internet. For example, the communication control unit 13 controls communication between the control unit 15 and a management device that manages operation logs, etc.

[0020] The storage unit 14 is implemented by a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 14 stores in advance a processing program for operating the similarity calculation device 10, data used during the execution of the processing program, etc., or temporarily stores them each time processing is performed. Note that the storage unit 14 may be configured to communicate with the control unit 15 via the communication control unit 13.

[0021] The control unit 15 is implemented using a CPU (Central Processing Unit) or the like, and executes a processing program stored in a memory. Thereby, as illustrated in FIG. 3, the control unit 15 functions as an acquisition unit 15a, a decomposition unit 15b, an extraction unit 15c, a determination unit 15d, and a calculation unit 15e, and executes a similarity calculation process. Note that these functional units may be implemented by different hardware, either individually or in part. For example, the calculation unit 15e may be implemented by hardware different from other functional units. Also, the control unit 15 may include other functional units.

[0022] The acquisition unit 15a acquires operation logs for a predetermined period of the processing target. For example, the acquisition unit 15a acquires the operation logs of the processing target via the input unit 11 or from a management device or the like via the communication control unit 13.

[0023] Note that the acquisition unit 15a may acquire the operation logs of the processing target in advance and store them in the storage unit 14, or may transfer them directly to the subsequent functional unit without storing them in the storage unit 14.

[0024] The decomposition unit 15b decomposes the URL included in the operation log to be processed into elements of the scheme syntax. Here, FIG. 4 is a diagram for explaining the processing of the decomposition unit. As illustrated in FIG. 4, the decomposition unit 15b decomposes each URL into scheme (the scheme), autorigy, pash (the path), query (the query), and fragment according to the http / https scheme syntax.

[0025] Here, the scheme is either http or https, which differs in the presence or absence of communication encryption, and can be ignored in calculating the similarity of the URLs. However, for example, the scheme may be used in calculating the similarity of the URLs, such as uniformly setting the similarity to 0 when the schemes are different.

[0026] The authority is the part representing the host name. When there is an exact match, the similarity is calculated, and when there is no exact match, the similarity is set to 0.

[0027] The authority may include user information and port numbers. Since differences in user information often do not affect the content of the web page, user information is ignored in calculating the similarity. When the port number is standard (80 for http and 443 for https), it is ignored in calculating the similarity. On the other hand, when the port number is non-standard, it is regarded as part of the host name and the similarity is calculated.

[0028] The fragment is the part representing the anchor within the web page. Since differences in the fragment do not affect the differences in the web page, it is ignored in calculating the similarity. However, the fragment may be considered in calculating the similarity.

[0029] Return to the description of FIG. 3. The extraction unit 15c extracts the part representing the ID from each of the two URLs to be processed included in the operation log. Specifically, the extraction unit 15c extracts the part representing the ID from the part constituting the path or query of each of the two URLs to be processed. That is, the extraction unit 15c extracts the part representing the ID from the path or query among the elements of the URL decomposed by the decomposition unit 15b. At that time, if percent encoding (URL encoding) is included, the extraction unit 15c performs decoding.

[0030] Here, FIGS. 5 and 6 are diagrams for explaining the processing of the extraction unit and the determination unit. FIG. 5 illustrates the process of extracting the part representing the ID from the part constituting the path of the URL. FIG. 6 illustrates the process of extracting the part representing the ID from the part constituting the query of the URL.

[0031] First, as illustrated in FIG. 5, the extraction unit 15c extracts the part representing the ID from the part constituting the path of the URL. Specifically, the extraction unit 15c extracts the partial character strings divided by " / " and their hierarchical positions from the front as illustrated in FIG. 5(b) from the part constituting the path of the URL illustrated in FIG. 5(a).

[0032] Next, as illustrated in FIG. 5(c), the extraction unit 15c calculates an ID determination score for each partial character string. Here, the ID determination score is the score obtained by scoring each partial character string according to a predetermined rule and a predetermined score distribution. When this ID determination score is, for example, equal to or higher than a predetermined threshold (0 in this embodiment), the extraction unit 15c determines that this partial character string represents some ID. Thereby, the extraction unit 15c determines each partial character string as ID or non-ID as illustrated in FIG. 5(d).

[0033] As rules regarding paths, for example, the following non-statistical rules and statistical rules are exemplified. First, as non-statistical rules, the following four are exemplified. Note that a dictionary of English words including inflected forms such as the past tense and proper nouns is prepared in advance corresponding to the following non-statistical rule (1).

[0034] (1) When an English word with four or more characters is included, subtract 10 points from the ID determination score (2) When numbers and alphabets are mixed, add 5 points to the ID determination score (3) When the number of characters is three or less, subtract 3 points from the ID determination score (4) When a half-width or full-width space is included, subtract 3 points from the ID determination score

[0035] Also, the statistical rule uses statistical information in the set of URLs included in all operation logs to determine whether the target partial string is an ID. For example, among the set of URLs included in all operation logs, statistical information in a subset of URLs where the authority string matches and the partial string at a hierarchical position higher than the hierarchical position of the path matches is used. That is, in this subset, when there are a plurality of candidates with the same string length for all partial strings of the partial string at the hierarchical position of the path, add 8 points to the ID determination score.

[0036] Also, as exemplified in FIG. 6, the extraction unit 15c extracts the part representing the ID from the part constituting the query of the URL. Specifically, the extraction unit 15c extracts a partial string representing the value of the key divided by "=" or "&" as exemplified in FIG. 6(b) from the part constituting the query of the URL exemplified in FIG. 6(a).

[0037] Here, there are two types of structures for the query string. Type 1 has a structure where keys and values are combined by '=', such as "key1=val1&key2=val2&···", and each key is connected by '&'. In this case, the extraction unit 15c extracts vol1 of key1, vol2 of key2, ···, and the position of the value is specified by the corresponding key.

[0038] Also, Type 2 has a structure where there are no keys and the values are connected by '&', such as "vol1&vol2&···". In this case, the extraction unit 15c extracts vol1, vol2, ···, and the position of the value is specified by the order. Note that when there is only one value, it is Type 2.

[0039] Next, as illustrated in FIG. 6(c), the extraction unit 15c calculates an ID determination score for each partial string. And when this ID determination score is, for example, equal to or greater than a predetermined threshold (0 in this embodiment), the extraction unit 15c determines that this partial string represents some ID. Thereby, as shown in FIG. 6(d), the extraction unit 15c determines each partial string as ID or non - ID.

[0040] The ID determination score is, similar to the ID determination score for the path, the score obtained by scoring each partial string according to a predetermined rule and a predetermined score distribution for the query. The rules for the query, similar to the rules for the path, include non - statistical rules and statistical rules. Among them, the non - statistical rules are the same as the non - statistical rules for the path.

[0041] Also, similar to the statistical rules for type 1 queries, the statistical rules for type 1 queries use statistical information in the set of URLs included in all operation logs to determine whether the target substring is an ID. For example, among the set of URLs included in all operation logs, the statistical information in the subset of URLs where the authority string and the path string match and that includes the key of the query is used. That is, in this subset, when there are multiple candidates where the string lengths of all substrings corresponding to the key are the same, 8 points are added to the ID determination score.

[0042] Similarly, the statistical rules for type 2 queries use, for example, the statistical information in the subset of URLs where the authority string and the path string match among the set of URLs included in all operation logs. That is, in this subset, when there are multiple candidates where the string lengths of all substrings corresponding to the position are the same, 8 points are added to the ID determination score.

[0043] Note that the extraction unit 15c may determine whether it is an ID using other parameters. Also, the determination process of whether it is an ID by the extraction unit 15c is not limited to the above. For example, the ID determination score may be obtained using only non-statistical rules. Alternatively, by performing the statistical rule process before the non-statistical rule process, the computational efficiency may be improved.

[0044] Returning to the description of FIG. 3, the determination unit 15d determines whether the part representing the ID is a temporarily generated part using the statistical information in the operation logs for a predetermined period. Here, a part representing an ID that is temporarily generated and discarded, such as a communication session ID, is defined as a temporary ID. Also, a part representing an ID that has a permanent meaning without changing depending on the access timing or the like is defined as a permanent ID.

[0045] The determination unit 15d determines whether it is a temporary ID or a permanent ID by using the statistical information in the set of URLs included in all operation logs. For example, the determination unit 15d determines whether it is a temporarily generated part by using the number of occurrences of the part representing the ID in the operation logs for a predetermined period. Specifically, when the part representing the ID appears two or more times in the operation logs for a predetermined period exceeding a predetermined time interval, the determination unit 15d determines that the part representing the ID is not a temporarily generated part.

[0046] For example, as illustrated in FIG. 5(e), the determination unit 15d determines whether the part representing the ID extracted from the part constituting the path of the URL is a temporary ID or a permanent ID. That is, the determination unit 15d uses the statistical information in the subset of URLs in the set of URLs included in all operation logs, where the authority strings match and the partial strings at hierarchical positions higher than the hierarchical position of the path match. For example, in this subset, when the same partial string representing the ID at the hierarchical position of the path appears two or more times at intervals of 12 hours or more, the determination unit 15d determines the partial string representing this ID as a permanent ID. In addition, the determination unit 15d determines the partial strings representing other IDs as temporary IDs.

[0047] In addition, as illustrated in FIG. 6(e), the determination unit 15d determines whether the part representing the ID extracted from the part constituting the query of the URL is a temporary ID or a permanent ID. That is, for type 1 queries, the determination unit 15d uses the statistical information in the subset of URLs in the set of URLs included in all operation logs, where the authority string and the path string match and the subset of URLs includes the key of the query. For example, in this subset, when the value corresponding to the key appears two or more times at intervals of 12 hours or more, the determination unit 15d determines the partial string representing this ID as a permanent ID. In addition, the determination unit 15d determines the partial strings representing other IDs as temporary IDs.

[0048] Further, for the type 2 query, the determination unit 15d uses the statistical information in the subset of URLs in the set of URLs included in all operation logs where the authority string and the path string match. For example, in this subset, if the value corresponding to the position appears two or more times at intervals of 12 hours or more, the determination unit 15d determines the partial string representing this ID as a persistent ID. In addition, the determination unit 15d determines the partial strings representing other IDs as temporary IDs.

[0049] Note that the determination unit 15d may determine the part representing the ID with two values such as a temporary ID / persistent ID as described above, or may determine it with a value having a width such as 0% to 100% (0.0 to 1.0).

[0050] Returning to the description of FIG. 3, when the part representing the ID is a temporarily generated part, the calculation unit 15e excludes the part representing the ID and calculates the similarity between the two URLs to be processed. In addition, when the part representing the ID is not a temporarily generated part, the calculation unit 15e adds a predetermined weight to the part representing the ID and calculates the similarity between the two URLs to be processed.

[0051] Specifically, first, when the authority parts of the two URLs to be processed do not exactly match, the calculation unit 15e sets the similarity to 0.

[0052] In addition, when the authority parts of the two URLs to be processed exactly match, the calculation unit 15e initializes the variables "similar points" and "maximum similar points" to 0, respectively.

[0053] Next, for the elements determined as non-IDs in the partial strings of the path / query, the calculation unit 15e compares them position by position, and adds 1 to "similar points" in the case of an exact match. In that case, when the partial string is not included at the position corresponding to one of the URLs, it is regarded as including a NULL string and is not considered an exact match.

[0054] Further, the calculation unit 15e adds the number of times of comparison for the elements determined to be non-IDs among the partial character strings of the path / query to the "maximum similarity point".

[0055] Next, the calculation unit 15e compares each element determined to be a persistent ID among the partial character strings of the path / query for each position, and adds 2 to the "similarity point" in the case of a perfect match. In that case, if the partial character string is not included at the position corresponding to one of the URLs, it is regarded as including a NULL character string and is considered not to be a perfect match.

[0056] Also, the calculation unit 15e weights the number of times of comparison for the elements determined to be persistent IDs among the partial character strings of the path / query by 2 and adds it to the "maximum similarity point".

[0057] Then, the calculation unit 15e calculates "similarity point" ÷ "maximum similarity point" as the similarity. In this way, the calculation unit 15e excludes the temporary ID from the comparison target of the similarity, adds a predetermined weight to the persistent ID, and calculates the similarity between the two URLs.

[0058] [Similarity calculation process] Next, with reference to FIG. 7, the similarity calculation process by the similarity calculation device 10 according to the present embodiment will be described. FIG. 7 is a flowchart showing the similarity calculation procedure. The flowchart of FIG. 7 starts, for example, at the timing when the user instructs the start of this device.

[0059] First, the acquisition unit 15a acquires the operation log for a predetermined period to be processed, and the decomposition unit 15b decomposes the URLs included in the operation log to be processed into elements of the scheme syntax. Further, the extraction unit 15c extracts the part representing the ID from each of the two URLs to be processed included in the operation log (step S1).

[0060] Specifically, the extraction unit 15c extracts the part representing the ID from each of the paths or queries of the two URLs to be processed. That is, the extraction unit 15c extracts the part representing the ID from the paths or queries among the elements of the URL decomposed by the decomposition unit 15b.

[0061] Next, the determination unit 15d determines whether the part representing the ID is a temporary ID temporarily generated or a permanent ID by using the statistical information in the set of URLs included in all the operation logs for a predetermined period (step S2).

[0062] For example, the determination unit 15d determines whether it is a temporary ID or a permanent ID by using the number of occurrences of the part representing the ID in the operation logs for a predetermined period. Specifically, the determination unit 15d determines that it is a permanent ID when the part representing the ID appears two or more times in the operation logs for a predetermined period exceeding a predetermined time interval.

[0063] Then, the calculation unit 15e excludes the temporary ID from the comparison targets of the similarity, adds a predetermined weight to the permanent ID, and calculates the similarity of the two URLs (step S3). Thereby, a series of similarity calculation processes is completed.

[0064] [Effect] As described above, in the similarity calculation device 10 of the present embodiment, the extraction unit 15c extracts the part representing the ID from each of the two URLs to be processed included in the operation logs. The determination unit 15d determines whether the part representing the ID is a part temporarily generated by using the statistical information in the operation logs for a predetermined period. When the part representing the ID is a part temporarily generated, the calculation unit 15e excludes the part representing the ID and calculates the similarity of the two URLs to be processed.

[0065] Specifically, the extraction unit 15c extracts the part representing the ID from each path or query that constitutes the two URLs to be processed. Further, the determination unit 15d determines whether it is a temporarily generated part by using the number of occurrences of the part representing the ID in the operation log for a predetermined period. For example, when the part representing the ID appears two or more times in the operation log for a predetermined period exceeding a predetermined time interval, the determination unit 15d determines that the part representing the ID is not a temporarily generated part.

[0066] In this way, the similarity calculation device 10 determines a temporary ID that has no influence on the similarity of the two URLs to be evaluated by using a non-statistical rule and a statistical rule, excludes this, and calculates the similarity of the two URLs. As a result, the similarity calculation device 10 can calculate the similarity of the URLs required for determining the identity of the web page with high accuracy.

[0067] In addition, when the part representing the ID is not a temporarily generated part, the calculation unit 15e further adds a predetermined weight to the part representing the ID to calculate the similarity of the two URLs to be processed. As a result, the similarity calculation device 10 can calculate the similarity of the URLs with even higher accuracy.

[0068] [Program] It is also possible to create a program that describes the processing executed by the similarity calculation device 10 according to the above embodiment in a language that can be executed by a computer. As an embodiment, the similarity calculation device 10 can be implemented by installing a similarity calculation program that executes the above similarity calculation process as package software or online software on a desired computer. For example, by causing the information processing device to execute the above similarity calculation program, the information processing device can function as the similarity calculation device 10. The information processing device mentioned here includes desktop or notebook personal computers. In addition, other information processing devices include mobile communication terminals such as smartphones, mobile phones, and PHS (Personal Handyphone System), and further slate terminals such as PDAs (Personal Digital Assistant) are included in this category. Also, the functions of the similarity calculation device 10 may be implemented on a cloud server.

[0069] FIG. 8 is a diagram showing an example of a computer that executes a similarity calculation program. The computer 1000 includes, for example, a memory 1010, a CPU 1020, a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These components are connected by a bus 1080.

[0070] Memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM 1012. The ROM 1011 stores a boot program such as a BIOS (Basic Input Output System), for example. The hard disk drive interface 1030 is connected to a hard disk drive 1031. The disk drive interface 1040 is connected to a disk drive 1041. A removable storage medium such as a magnetic disk or an optical disk is inserted into the disk drive 1041, for example. A mouse 1051 and a keyboard 1052 are connected to the serial port interface 1050, for example. A display 1061 is connected to the video adapter 1060, for example.

[0071] Here, the hard disk drive 1031 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. Each piece of information described in the above embodiment is stored in, for example, the hard disk drive 1031 or the memory 1010.

[0072] Also, the similarity calculation program is stored in the hard disk drive 1031 as a program module 1093 in which instructions executed by a computer 1000 are described, for example. Specifically, a program module 1093 in which each process executed by the similarity calculation device 10 described in the above embodiment is described is stored in the hard disk drive 1031.

[0073] Also, data used for information processing by the similarity calculation program is stored in the hard disk drive 1031 as program data 1094, for example. Then, the CPU 1020 reads out the program module 1093 and the program data 1094 stored in the hard disk drive 1031 into the RAM 1012 as needed, and executes each of the above-described procedures.

[0074] Note that the program module 1093 and program data 1094 related to the similarity calculation program are not limited to being stored in the hard disk drive 1031. For example, they may be stored in a removable storage medium and read by the CPU 1020 via a disk drive 1041 or the like. Alternatively, the program module 1093 and program data 1094 related to the similarity calculation program may be stored in another computer connected via a network such as a LAN or WAN (Wide Area Network), and read by the CPU 1020 via the network interface 1070.

[0075] As described above, the embodiments to which the invention made by the present inventor is applied have been described. However, the present invention is not limited by the description and drawings that form a part of the disclosure of the present invention according to the present embodiment. That is, all other embodiments, examples, operation techniques, etc. made by those skilled in the art based on the present embodiment are included in the scope of the present invention.

Explanation of Reference Numerals

[0076] 10 Similarity calculation device 11 Input unit 12 Output unit 13 Communication control unit 14 Storage unit 15 Control unit 15a Acquisition unit 15b Decomposition unit 15c Extraction unit 15d Determination unit 15e Calculation unit

Claims

1. An extraction unit that extracts a portion representing an ID from each of two URLs to be processed included in an operation log; A determination unit that determines whether or not the portion representing the ID is a portion that was temporarily generated, using statistical information in the operation log for a predetermined period; A calculation unit that, when the portion representing the ID is a portion that was temporarily generated, excludes the portion representing the ID and calculates the similarity between the two URLs to be processed; A similarity calculation device, characterized by comprising the above.

2. The similarity calculation device according to claim 1, wherein the determination unit determines whether or not it is a portion that was temporarily generated, using the number of occurrences of the portion representing the ID in the operation log for a predetermined period.

3. The similarity calculation device according to claim 2, wherein the determination unit determines that the portion representing the ID is not a portion that was temporarily generated when the portion representing the ID appears two or more times in the operation log for a predetermined period exceeding a predetermined time interval.

4. The similarity calculation device according to claim 1, wherein when the portion representing the ID is not a portion that was temporarily generated, the calculation unit further adds a predetermined weight to the portion representing the ID and calculates the similarity between the two URLs to be processed.

5. The similarity calculation device according to claim 1, wherein the extraction unit extracts the portion representing the ID from a portion constituting each path or query of the two URLs to be processed.

6. A similarity calculation method executed by a similarity calculation device, comprising: An extraction step of extracting a portion representing an ID from each of two URLs to be processed included in an operation log; A determination step of determining whether or not the portion representing the ID is a portion that was temporarily generated, using statistical information in the operation log for a predetermined period; A calculation step of excluding the portion representing the ID and calculating the similarity between the two URLs to be processed when the portion representing the ID is a portion that was temporarily generated; A similarity calculation method, characterized by including the above.

7. An extraction step of extracting a portion representing an ID from each of two URLs to be processed included in an operation log; A determination step of determining whether or not the portion representing the ID is a portion that was temporarily generated, using statistical information in the operation log for a predetermined period; When the part representing the ID is a temporarily generated part, a calculation step of calculating the similarity of the two URLs to be processed by excluding the part representing the ID; A similarity calculation program characterized by causing a computer to execute the program.

Images