Information management device, information management method, and information management program

Abstract

A related word acquisition unit (212) acquires, from a vocabulary list (32) in which a plurality of words representing similar targets are associated with each other, a related word, which is associated with an input word that has been input. A search unit (213) uses the input word to search management information and uses the related word acquired by the related word acquisition unit (212) to search management information (31), thereby identifying a retrieval target. For the retrieval target identified by the search unit (213), an evaluation value calculation unit (214) calculates an evaluation value from the degree of similarity between a word used during search for the retrieval target and the input word.

Description

Technical Field

[0001] The present disclosure relates to a technique for identifying a search target in consideration of differences in the expression of words used in a search.

Background Art

[0002] The construction of smart cities is underway. In the provision of smart city services, the management of assets across multiple fields is performed using a city OS. OS is an abbreviation for Operating System. In the provision of smart city services, there are situations where it is desired to identify a target asset.

[0003] Patent Document 1 describes generating a knowledge graph by associating spatial data, facility data, and measurement data according to ontology conversion rules, and searching the knowledge graph with input search information. Thus, in Patent Document 1, the spatial data, facility data, and measurement data corresponding to the search information are identified.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] When assets across multiple fields are being managed, it may be considered to use the technology of Patent Document 1 to perform a search for a target asset using a knowledge graph. However, there are differences in vocabulary expressions depending on the field. Therefore, if the user does not fully understand the assets being managed, it is difficult to search for the assets with appropriate keywords. If the search is not performed with appropriate keywords, there is a possibility that the target asset will not be extracted. This disclosure aims to enable the appropriate identification of target search results by searching across different fields. [Means for solving the problem]

[0006] The information management device related to this disclosure is A related word acquisition unit retrieves related words that are associated with the input word from a vocabulary list in which multiple words representing similar objects are associated, A search unit that identifies a search target by performing a search on the management information using the input word and by performing a search on the management information using the related word obtained by the related word acquisition unit, An evaluation value calculation unit calculates an evaluation value for the search target identified by the search unit, based on the similarity between the words used during the search of the search target and the input words. It is equipped with. [Effects of the Invention]

[0007] This disclosure identifies the search target by performing a search using the input word and also by performing a search using related words associated with the input word. Then, for the search target, an evaluation value is calculated from the similarity between the words used during the search and the input word. By performing a search using not only the input word but also related words, it is possible to appropriately extract the search target even when searching across different fields. Furthermore, by referring to the evaluation value, it is possible to appropriately identify the target search. [Brief explanation of the drawing]

[0008] [Figure 1] Hardware configuration diagram of the information management device 10 according to Embodiment 1. [Figure 2] Functional configuration diagram of the information management device 10 according to Embodiment 1. [Figure 3] An explanatory diagram of the management information 31 according to Embodiment 1. [Figure 4] An explanatory diagram of the vocabulary list 32 according to Embodiment 1. [Figure 5] Flowchart of the processing of the information management device 10 according to Embodiment 1. [Figure 6] Explanatory diagram of the word input processing according to Embodiment 1. [Figure 7] Explanatory diagram of the result output processing according to Embodiment 1. [Figure 8] Explanatory diagram of another example according to Embodiment 1. [Figure 9] Explanatory diagram of another example according to Embodiment 1. [Figure 10] Functional configuration diagram of the information management device 10 according to Embodiment 2. [Figure 11] Explanatory diagram of the correction rule 33 according to Embodiment 2. [Figure 12] Flowchart of the processing of the information management device 10 according to Embodiment 2. [Figure 13] Explanatory diagram of the vocabulary list 32 according to Embodiment 2. [Figure 14] Explanatory diagram of the management information 31 according to Embodiment 2. [Figure 15] Explanatory diagram of the knowledge graph according to Embodiment 2. [Figure 16] Functional configuration diagram of the information management device 10 according to Embodiment 3. [Figure 17] Explanatory diagram of the affiliation information 34 according to Embodiment 3. [Figure 18] Explanatory diagram of the vocabulary list 32 according to Embodiment 3. [Figure 19] Flowchart of the processing of the information management device 10 according to Embodiment 3. [Figure 20] Functional configuration diagram of the information management device 10 according to Embodiment 4. [Figure 21] Flowchart of the processing of the information management device 10 according to Embodiment 4. [Figure 22] Explanatory diagram of the data regarding the search target according to Embodiment 4. [Figure 23] Functional configuration diagram of the information management device 10 according to Embodiment 5. [Figure 24] Flowchart of the processing of the information management device 10 according to Embodiment 5. [Modes for carrying out the invention]

[0009] Embodiment 1. Embodiment 1 describes an information management device 10 that manages assets across multiple fields as search targets.

[0010] ***Explanation of the structure*** Referring to Figure 1, the hardware configuration of the information management device 10 according to Embodiment 1 will be described. The information management device 10 is a computer. The information management device 10 comprises hardware including a processor 11, memory 12, storage 13, and a communication interface 14. The processor 11 is connected to the other hardware via signal lines and controls this other hardware.

[0011] Processor 11 is an IC that performs processing. IC stands for Integrated Circuit. Specific examples of processor 11 include CPU, DSP, and GPU. CPU stands for Central Processing Unit. DSP stands for Digital Signal Processor. GPU stands for Graphics Processing Unit.

[0012] Memory 12 is a storage device that temporarily stores data. Specific examples of memory 12 include SRAM and DRAM. SRAM stands for Static Random Access Memory. DRAM stands for Dynamic Random Access Memory.

[0013] Storage 13 is a storage device for storing data. A concrete example of storage 13 is an HDD. HDD stands for Hard Disk Drive. Alternatively, storage 13 may be a portable recording medium such as an SD® memory card, CompactFlash®, NAND flash, flexible disk, optical disk, compact disk, Blu-ray® disc, or DVD. SD stands for Secure Digital. DVD stands for Digital Versatile Disk.

[0014] Communication interface 14 is an interface for communicating with external devices. Specific examples of communication interface 14 include Ethernet®, USB, and HDMI® ports. USB stands for Universal Serial Bus. HDMI stands for High-Definition Multimedia Interface.

[0015] Referring to Figure 2, the functional configuration of the information management device 10 according to Embodiment 1 will be described. The information management device 10 comprises, as functional components, a target management unit 21, a data management unit 22, and a communication processing unit 23. The target management unit 21 has the function of searching for assets that are the target of the search. The data management unit has the function of managing information about the assets that are the target of the search. The communication processing unit has the function of controlling communication between the system related to the assets that are the target of the search and the application program that inputs the search instructions. The target management unit 21 comprises functional components including a word input unit 211, a related word acquisition unit 212, a search unit 213, an evaluation value calculation unit 214, and a result output unit 215. The functions of each functional component of the information management device 10 are realized by software. The storage 13 stores programs that implement the functions of each functional component of the information management device 10. These programs are loaded into memory 12 by the processor 11 and executed by the processor 11. This realizes the functions of each functional component of the information management device 10.

[0016] Storage 13 stores management information 31 and a vocabulary list 32. Management information 31 is information about the asset to be searched. Vocabulary list 32 is information that associates multiple words that represent similar objects.

[0017] In Figure 1, only one processor 11 was shown. However, there may be multiple processors 11, and multiple processors 11 may work together to execute programs that implement each function.

[0018] ***Explanation of operation*** Referring to Figures 3 to 8, the operation of the information management device 10 according to Embodiment 1 will be explained. The operating procedure of the information management device 10 according to Embodiment 1 corresponds to the information management method according to Embodiment 1. Furthermore, the program that implements the operation of the information management device 10 according to Embodiment 1 corresponds to the information management program according to Embodiment 1.

[0019] Referring to Figure 3, the management information 31 according to Embodiment 1 will be described. Management information 31 contains words for each of the multiple classifications for each asset. Here, we will explain using an example where management information 31 contains words for each of the classifications of name (Entity), function (Relationship), and attribute (Property) in the knowledge graph. Note that classifications are not limited to name, function, and attribute in the knowledge graph. Classifications may also be, for example, types of multiple attributes. Furthermore, it is not necessary for words to be set for all classifications; words may be set for only some classifications. In Figure 3, for an asset named "Temperature Sensor," the function is set to "Measurement" and the attribute is set to "Temperature." Similarly, for an asset named "Air Conditioning Indoor Unit," the function is set to "Acquisition" and the attributes are set to "Room Temperature" and "Humidity."

[0020] Referring to Figure 4, the vocabulary list 32 according to Embodiment 1 will be described. Vocabulary list 32 associates multiple words that represent similar subjects, and assigns a degree of similarity between these words. Different fields may use different words to represent similar subjects. Vocabulary list 32 associates these words that are expressed differently depending on the field. In Figure 4, the words "acquisition" and "measurement" are associated, with a similarity score of 0.9. Similarly, the words "temperature" and "room temperature" are associated, with a similarity score of 0.8. Here, a similarity score closer to 1 indicates greater similarity. The similarity score between identical words is set to 1.

[0021] Referring to Figure 5, the processing of the information management device 10 according to Embodiment 1 will be explained. (Step S11: Word input processing) The word input unit 211 accepts input words for searching for assets. The accepted words are called input words. Specifically, the word input unit 211 displays an input screen as shown in Figure 6, allowing the user to input words for each category. In other words, the word input unit 211 accepts word input for each of the categories: name, function, and attribute. Note that it is not necessary to input words for all categories; words may be entered for only some of the categories. In Figure 6, the word "temperature" has been entered for the attribute category.

[0022] (Step S12: Related word acquisition process) The related word acquisition unit 212 reads the vocabulary list 32. Then, the related word acquisition unit 212 acquires the words associated with the input word entered in step S11 from the vocabulary list 32 as related words. At this time, the related word acquisition unit 212 also acquires the similarity between the input word and the related words set in the vocabulary list 32. For example, suppose the input word is "temperature," as shown in Figure 6. In the vocabulary list 32 shown in Figure 4, "room temperature" is set as a word related to "temperature." Therefore, the related word acquisition unit 212 acquires "room temperature" as a related word when "temperature" is the input word. At this time, the related word acquisition unit 212 also acquires a similarity score of 0.8 between "temperature" and "room temperature" set in the vocabulary list 32.

[0023] (Step S13: Search process) The search unit 213 performs a search on the management information 31 using the input word entered in step S11. The search unit 213 also performs a search on the management information 31 using the related word obtained in step S12. This allows the search unit 213 to identify the asset that corresponds to the input word or related word. For example, suppose "temperature" is the input word and "room temperature" is the related word. In this case, the search unit 213 searches the attribute classification of the management information 31 using "temperature" as the keyword. This identifies the temperature sensor from the management information 31 shown in Figure 3. The search unit 213 also searches the attribute classification of the management information 31 using "room temperature" as the keyword. This identifies the air conditioning indoor unit from the management information 31 shown in Figure 3. In other words, the temperature sensor and the air conditioning indoor unit are identified.

[0024] (Step S14: Evaluation value calculation process) The evaluation value calculation unit 214 calculates an evaluation value for each asset identified in step S13. Specifically, the evaluation value calculation unit 214 sets each asset as a target asset. The evaluation value calculation unit 214 calculates an evaluation value from the similarity between the words used when searching for the target asset and the input words entered in step S11. As a specific example, the evaluation value calculation unit 214 uses the similarity directly as the evaluation value. For example, a temperature sensor was searched using "temperature" as the keyword. Since "temperature" is an input word, the similarity with the input word is 1. Therefore, the evaluation value for the temperature sensor is 1. Also, an air conditioning indoor unit was searched using "room temperature" as the keyword. "Room temperature" is a related word, and the similarity with the input word "temperature" is 0.8, as obtained in step S12. Therefore, the evaluation value for the air conditioning indoor unit is 0.8.

[0025] (Step S15: Result output processing) The result output unit 215 outputs the assets identified in step S13 along with the evaluation values ​​calculated in step S14. For example, if temperature is input as an input word as shown in Figure 6, the result output unit 215 outputs the temperature sensor and the air conditioning indoor unit as assets, as shown in Figure 7. In this case, the result output unit 215 outputs 1 as the evaluation value for the temperature sensor and 0.8 as the evaluation value for the air conditioning indoor unit.

[0026] As shown in Figure 8, when "acquisition" is entered as the function classification in step S11, in step S12 the related word acquisition unit 212 acquires "measurement" as a related word for "acquisition" from the vocabulary list 32 shown in Figure 4. Then, in step S13 the search unit 213 searches for the functions of the management information 31 using "acquisition" and "measurement" as keywords. By searching with "acquisition" as the keyword, the air conditioning indoor unit is identified. By searching with "measurement" as the keyword, the temperature sensor is identified. In step S14 the evaluation value calculation unit 214 calculates evaluation values ​​for both the air conditioning indoor unit and the temperature sensor. The air conditioning indoor unit was searched using the input word "acquisition" as the keyword, so the evaluation value is 1. The temperature sensor was searched using the related word "measurement" as the keyword, and since the similarity between "measurement" and the input word "acquisition" is 0.9, the evaluation value is 0.9.

[0027] In the example in Figure 6, words are entered only in the attribute field in step S11, while in the example in Figure 8, words are entered only in the function field in step S11. However, words may be entered in two or more classification fields. In this case, in step S12, the related word acquisition unit 212 acquires related words for each input word entered for two or more classifications. In step S13, the search unit 213 identifies the asset to be searched by performing a search on the management information 31 using each combination obtained by setting the input word or related word for each of the two or more classifications. In step S14, the evaluation value calculation unit 214 calculates an evaluation value for each asset as a target asset, based on the similarity between each word included in the combination used when searching for the target asset and the input word. Specifically, the evaluation value calculation unit 214 calculates the evaluation value as the product of the similarity between each word included in the combination used when searching and the input word.

[0028] For example, as shown in Figure 9, suppose that in step S11, "acquisition" is entered for the function classification and "temperature" is entered for the attribute classification. In this case, in step S12, the related word acquisition unit 212 acquires "measurement" as the related word to acquire from the vocabulary list 32 shown in Figure 4. It also acquires "room temperature" as the related word for temperature. Here, the related word acquisition unit 212 acquires 0.9 as the similarity of "measurement" and 0.8 as the similarity of "room temperature".

[0029] In step S13, the search unit 213 performs a search on the management information 31 using each combination in which the function classification is set to acquisition or measurement and the attribute classification is set to temperature or room temperature. That is, it performs a search on the management information 31 using each of the following combinations 1 to 4. Combination 1 is a combination in which the function classification is set to acquisition and the attribute classification is set to temperature. Combination 2 is a combination in which the function classification is set to acquisition and the attribute classification is set to room temperature. Combination 3 is a combination in which the function classification is set to measurement and the attribute classification is set to temperature. Combination 4 is a combination in which the function classification is set to measurement and the attribute classification is set to room temperature. Here, performing a search on management information 31 using combination 1 means searching for assets where the function is set to "Get" and the attribute is set to "Temperature". The same applies to combinations 2 through 4. Then, using the management information 31 shown in Figure 3, the indoor air conditioning unit is identified by searching using combination 2, and the temperature sensor is identified by searching using combination 3.

[0030] In step S14, the evaluation value calculation unit 214 calculates evaluation values ​​for both the air conditioning indoor unit and the temperature sensor. The calculation for the indoor unit of the air conditioner is as follows: The acquired item in combination 2 is the input word with a similarity of 1, and the room temperature is the related word with a similarity of 0.8. Therefore, the evaluation value calculation unit 214 calculates the evaluation value as 0.8 by multiplying 1 by 0.8. For the temperature sensor, the calculation is performed as follows: The measurement included in combination 3 is a related word with a similarity of 0.9, and temperature is an input word with a similarity of 1. Therefore, the evaluation value calculation unit 214 calculates the evaluation value as 0.9 × 1, resulting in 0.9.

[0031] ***Effects of Embodiment 1*** As described above, the information management device 10 according to Embodiment 1 identifies the search target by performing a search using related words associated with the input word. Then, it calculates an evaluation value for the search target from the similarity between the words used during the search and the input word. There may be differences in vocabulary and expression depending on the field. Even when searching across different fields with differing vocabulary and expression, it is possible to appropriately extract the search target by using not only the input word but also related words. Furthermore, by referring to the evaluation value, it is possible to appropriately identify the desired search target.

[0032] ***Other configurations*** <Example 1> In Embodiment 1, each functional component was implemented in software. However, in Modification 1, each functional component may be implemented in hardware. The differences between this Modification 1 and Embodiment 1 will be explained below.

[0033] When each functional component is implemented in hardware, the information management device 10 includes electronic circuits instead of the processor 11, memory 12, and storage 13. The electronic circuits are dedicated circuits that implement the functions of each functional component, as well as the functions of the memory 12 and storage 13.

[0034] Electronic circuits can include single circuits, complex circuits, programmed processors, parallel programmed processors, logic ICs, GAs, ASICs, and FPGAs. GA stands for Gate Array. ASIC stands for Application Specific Integrated Circuit. FPGA stands for Field-Programmable Gate Array. Each functional component may be implemented in a single electronic circuit, or it may be implemented by distributing each functional component across multiple electronic circuits.

[0035] <Modification 2> As a second variation, some of the functional components may be implemented in hardware, while others may be implemented in software.

[0036] The processor 11, memory 12, storage 13, and electronic circuitry are collectively referred to as the processing circuit. In other words, the function of each functional component is realized by the processing circuit.

[0037] Furthermore, the term "part" in the above explanation may be replaced with "circuit," "process," "procedure," "processing," or "processing circuit."

[0038] Embodiment 2. Embodiment 2 differs from Embodiment 1 in that the evaluation value is calculated using the structure of the knowledge graph about the asset being searched. Embodiment 2 explains this difference, while omitting explanations of the same points.

[0039] ***Explanation of the structure*** Referring to Figure 10, the functional configuration of the information management device 10 according to Embodiment 2 will be described. Information management device 10 differs from the information management device 10 shown in Figure 2 in that its evaluation value calculation unit 214 includes a similarity calculation unit 216, a structure comparison unit 217, and a correction unit 218. Furthermore, information management device 10 differs from the information management device 10 shown in Figure 2 in that correction rules 33 are stored in the storage unit 13. The correction rules 33 represent rules for correcting evaluation values ​​based on the structure of the knowledge graph.

[0040] ***Explanation of operation*** The correction rule 33 according to Embodiment 2 will be explained with reference to Figure 11. Correction Rule 33 outlines the rules for correcting evaluation values. Specifically, correction rule 33 is a rule that corrects the evaluation value based on the degree of agreement between the structure of the knowledge graph of the target asset identified by searching using the input word and the structure of the knowledge graph of the target asset identified by searching using related words. Here, correction rule 33 shows a rule that depends on how many hops away the structure matches from the identified asset. Correction rule 33 is a rule that corrects the evaluation value so that it approaches 1 the further away the structure matches.

[0041] Referring to Figure 12, the processing of the information management device 10 according to Embodiment 2 will be explained. The processes from steps S21 to S23 are the same as the processes from steps S11 to S13 in Figure 5. Also, the process in step S28 is the same as the process in step S15 in Figure 5.

[0042] The processes from step S24 to step S27 correspond to the evaluation value calculation process in step S14 of Figure 5. (Step S24: Similarity calculation process) The similarity calculation unit 216 calculates an evaluation value for each asset identified in step S23, similar to step S14 in Figure 5, based on the similarity between the words used when searching for that asset and the input words entered in step S21.

[0043] (Step S25: Structural comparison process) The structure comparison unit 217 compares the structure of the input graph with the structure of the related graph to determine the degree of similarity. The input graph is a knowledge graph about the asset that is the target of the search and has been identified by performing a search using the input word. The related graph is a knowledge graph about the asset that is the target of the search and has been identified by performing a search using the related word. Here, the structural comparison unit 217 identifies the degree of similarity, which is the number of hops away from the identified asset where the structure matches.

[0044] This will be explained in detail with reference to Figures 13 to 15. As shown in Figure 13, the vocabulary list 32 is assumed to associate the words "point data" and "monitoring data". In this case, in step S21, the word "point data" is specified for attribute classification. Then, in step S22, monitoring data is retrieved as an associated word for point data. Then, in step S23, a search is performed using point data and monitoring data as keywords. As a result, from the management information 31 shown in Figure 14, if point data is searched using the keyword IoTGW is identified, and if monitoring data is searched using the keyword remote monitoring device is identified. In step S24, the similarity calculation unit 216 calculates an evaluation value for both the IoTGW and the remote monitoring device based on the similarity between the input word (point data) and the word used during the search. Here, for the IoTGW, the evaluation value is 1 because the input word (point data) was used as a keyword in the search, and for the remote monitoring device, the evaluation value is 0.7 because the similarity between the monitoring data and the input word (point data) is 0.7.

[0045] In step S25, the structure comparison unit 217 compares the structure of the input graph with the structure of the related graph to determine the degree of agreement. First, the structural comparison unit 217 sets the assets found by searching using related words as keywords as target assets. In this case, only the remote monitoring device is set as the target asset. The structure comparison unit 217 compares the structure of the input graph, which is a knowledge graph about the IoTGW, with the structure of the relationship graph, which is a knowledge graph about the remote monitoring device set for the target asset. When the assets related to the IoTGW and the assets related to the remote monitoring device in the management information 31 shown in Figure 14 are represented as graph structures, the structure shown in Figure 15 is obtained. The structure comparison unit 217 determines the degree of agreement, which indicates how many hops away the structure of the input graph and the structure of the relationship graph match from the identified assets (in this case, the IoTGW and the remote monitoring device). Whether or not the structures match is determined based on whether the shape of the graphs is the same and whether the vocabulary of the functions (Relationship) is the same. If both are the same, the configuration is determined to be a perfect match, and if only one matches, the configuration is determined to be a partial match. In this case, the structure comparison unit 217 determines that the configuration is a perfect match up to two hops away. In other words, the degree of agreement is determined to be a perfect match up to two hops away.

[0046] (Step S26: Matching degree determination process) The structure comparison unit 217 determines in step S25 whether the structure of the input graph and the structure of the related graph match at least partially. The structural comparison unit 217 determines that correction of the evaluation value is necessary if there is at least a partial match, and proceeds to step S27. On the other hand, if there is no match at all, the structural comparison unit 217 determines that correction of the evaluation value is unnecessary, and proceeds to step S28.

[0047] (Step S27: Correction process) The correction unit 218 corrects the evaluation value calculated in step S24 according to the rule corresponding to the degree of agreement identified in step S25. In the examples in Figures 13 to 15, the degree of agreement is that the configurations are a perfect match up to two hops away. Therefore, the correction unit 218 reads the rule for cases where the configurations are a perfect match up to two hops away from the correction rule 33. As shown in Figure 11, the rule for cases where the configurations are a perfect match up to two hops away is: Corrected evaluation value = (evaluation value + 2) / 3. Therefore, the correction unit 218 substitutes the evaluation value calculated in step S24 into this formula. The evaluation value calculated in step S24 is 0.7. Thus, the corrected evaluation value becomes (0.7 + 2) / 3 = 0.9.

[0048] Therefore, in step S28, the result output unit 215 outputs IoTGW with an evaluation value of 1 and remote monitoring device with an evaluation value of 0.9.

[0049] ***Effects of Embodiment 2*** As described above, the information management device 10 according to Embodiment 2 calculates an evaluation value using the structure of the knowledge graph for the asset to be searched. Specifically, the information management device 10 calculates an evaluation value based on the degree of agreement between the structure of the input graph and the structure of the related graph. This makes it possible to calculate the evaluation value more appropriately.

[0050] Embodiment 3. Embodiment 3 differs from Embodiments 1 and 2 in that it limits related words based on the field to which the input word belongs. Embodiment 3 explains this difference, while omitting explanations of the same points. Embodiment 3 describes a case where functionality is added to Embodiment 1. However, it is also possible to add functionality to Embodiment 2.

[0051] ***Explanation of the structure*** Referring to Figure 16, the functional configuration of the information management device 10 according to Embodiment 3 will be described. The information management device 10 differs from the information management device 10 shown in Figure 2 in that the affiliation information 34 is stored in the storage 13. The affiliation information 34 indicates the field to which the input word belongs.

[0052] ***Explanation of operation*** Referring to Figure 17, the affiliation information 34 related to Embodiment 3 will be explained. Affiliation information 34 indicates the application program or user's field of expertise from which the input word originated. In Figure 17, assuming the input source is an application program, affiliation information 34 is set to include the fields of expertise for elevator programs and timetable search programs.

[0053] Referring to Figure 18, the vocabulary list 32 according to Embodiment 3 will be described. The vocabulary list 32 associates multiple words that represent similar objects, and assigns a similarity score to each word. Furthermore, the vocabulary list 32 assigns a usage field to each word. For example, in Figure 18, "landing" and "destination station" are associated, with a similarity score of 0.7. Additionally, the usage field for "landing" is "buildings," and the usage field for "destination station" is "transportation."

[0054] Referring to Figure 19, the processing of the information management device 10 according to Embodiment 3 will be explained. The processes in steps S31 and S32 are the same as the processes in steps S11 and S12 in Figure 5. The processes in steps S35 to S37 are the same as the processes in steps S13 to S15 in Figure 5.

[0055] The processes in steps S33 and S34 are executed for each of the one or more related words obtained in step S32.

[0056] (Step S33: Affiliation determination process) The related word acquisition unit 212 determines whether the field of use of the target related word is the same as the field to which the input word originates. Here, it is assumed that the application program or user from which the word was input is identified. For example, when accessing the information management device 10, authentication processing is performed, and the application program or user from which the word was input is identified. The related word acquisition unit 212 proceeds to step S34 if the field of use is the same as the field of belonging. On the other hand, if the field of use is different from the field of belonging, the related word acquisition unit 212 skips step S34 and terminates processing for the target related word.

[0057] (Step S34: Word exclusion process) The related word acquisition unit 212 excludes the target related word from the related words acquired in step S32. In other words, the related word acquisition unit 212 ensures that the target related word is not used as a keyword during the search in step S35. For example, suppose the destination station is input by the elevator program. Then, in step S32, "landing" is obtained as a related word. However, since the field to which the elevator program belongs and the field in which "landing" is used are the same building, the process proceeds to step S34 in step S33, and "landing" is excluded from the related words.

[0058] Thus, through the processing from step S32 to step S34, the related word acquisition unit 212 acquires as related words only those words whose registered usage field is different from the field to which the input word belongs. In other words, the related word acquisition unit 212 stops acquiring as related words words that have the same registered usage field as the field to which the input word belongs.

[0059] ***Effects of Embodiment 3*** As described above, the information management device 10 according to Embodiment 3 stops acquiring related words for words whose registered usage field is the same as the field to which the input word belongs. This reduces the number of related words. As a result, the processing load related to searching and calculating evaluation values ​​can be reduced. In addition, it is possible to prevent unintended search targets from being extracted.

[0060] Here, it is assumed that the vocabulary related to the user's field of study is appropriately understood. Therefore, the need to retrieve words related to the field of study as related words is considered low. For example, if the field of study is buildings, it is unlikely that the user would refer to "landing" as "landing station." Therefore, if "landing station" is entered, the need to retrieve "landing" as a related word and perform a search using "landing" is considered low. In other words, a search using "landing" as a keyword is unlikely to yield the search results intended by the user.

[0061] Embodiment 4. Embodiment 4 differs from Embodiments 1 to 3 in that it calculates evaluation values ​​using data about the search target. Embodiment 4 explains this difference, while omitting explanations of the same points. Embodiment 4 describes a case where functionality is added to Embodiment 1. However, it is also possible to add functionality to Embodiments 2 and 3.

[0062] ***Explanation of the structure*** Referring to Figure 20, the functional configuration of the information management device 10 according to Embodiment 4 will be described. The information management device 10 differs from the information management device 10 shown in Figure 2 in that its evaluation value calculation unit 214 includes a similarity calculation unit 216, a correction unit 218, and a data comparison unit 219.

[0063] ***Explanation of operation*** Referring to Figure 21, the processing of the information management device 10 according to Embodiment 4 will be explained. The processes from step S41 to step S43 are the same as the processes from step S11 to step S13 in Figure 5. Also, the process in step S48 is the same as the process in step S15 in Figure 5.

[0064] The processes from step S44 to step S47 correspond to the evaluation value calculation process in step S14 of Figure 5. (Step S44: Similarity calculation process) The similarity calculation unit 216 calculates an evaluation value for each asset identified in step S23, similar to step S14 in Figure 5, based on the similarity between the words used when searching for that asset and the input words entered in step S41.

[0065] (Step S45: Data acquisition process) The data comparison unit 219 retrieves data about the asset that is the search target identified in step S43. Information about the asset is managed by the data management unit 22. Specifically, the data comparison unit 219 performs a search using the input word and retrieves data about the input word related to the search target identified. The data comparison unit 219 also performs a search using related words and retrieves data about related words related to the search target identified. For example, as shown in Figure 22, suppose the input word is temperature and the related word is room temperature. Suppose a boiler is identified using temperature as the keyword, and an air conditioning indoor unit is identified using room temperature as the keyword. In this case, the data comparison unit 219 obtains the time-series values ​​of the boiler's temperature as data related to the boiler's temperature. The data comparison unit 219 also obtains the time-series values ​​of the air conditioning indoor unit's room temperature as data related to the air conditioning indoor unit's room temperature.

[0066] (Step S46: Data comparison processing) The data comparison unit 219 performs a search using the input word and compares the data relating to the input word for the search target identified with the data relating to the related word for the search target identified using the related word. The data comparison unit 219 then determines whether or not there is similarity between the data. In the example shown in Figure 22, the boiler temperature range is 25-65°C. In contrast, the room temperature range of the air conditioning indoor unit is 25-26°C. Therefore, it is determined that there is no similarity between the data. The data comparison unit 219 determines that if there is no similarity, correction of the evaluation value is necessary and proceeds to step S47. On the other hand, if there is similarity, the data comparison unit 219 determines that correction of the evaluation value is unnecessary and proceeds to step S48.

[0067] (Step S47: Correction process) The correction unit 218 corrects the evaluation value calculated in step S44 to a lower value. The correction unit 218 may also correct the evaluation value so that the lower the similarity, the lower the evaluation value. Similar to Embodiment 2, correction rules 33 indicating rules for each similarity may be stored in the storage 13 and the correction may be made by referring to them.

[0068] ***Effects of Embodiment 4*** As described above, the information management device 10 according to Embodiment 4 calculates an evaluation value using data about the search target. Even if a search target has related words set, if the search target handles different types of data, it is highly likely that it is not the intended search target. The information management device 10 according to Embodiment 4 makes it easier to identify the intended search target by lowering the evaluation value for such search targets.

[0069] Embodiment 5. Embodiment 5 differs from Embodiments 1 to 4 in that it displays related words along with their similarity scores before the search. Embodiment 5 explains this difference, while omitting explanations of the same points. Embodiment 5 describes a case where functionality is added to Embodiment 1. However, it is also possible to add functionality to Embodiments 2 to 4.

[0070] ***Explanation of the structure*** Referring to Figure 23, the functional configuration of the information management device 10 according to Embodiment 5 will be described. The information management device 10 differs from the information management device 10 shown in Figure 2 in that it includes a candidate display unit 220 as a functional component.

[0071] ***Explanation of operation*** Referring to Figure 24, the processing of the information management device 10 according to Embodiment 5 will be explained. The process in step S51 is the same as the process in step S11 in Figure 5. The processes in steps S55 to S58 are the same as the processes in steps S12 to S15 in Figure 5.

[0072] (Step S52: Candidate acquisition process) The related word acquisition unit 212 acquires related words from the vocabulary list 32, similar to step S12 in Figure 5, that are associated with the input word entered in step S11. At this time, the related word acquisition unit 212 also acquires the similarity between the input word set in the vocabulary list 32 and the related words.

[0073] (Step S53: Candidate display process) The candidate display unit 220 displays the input word entered in step S51 and the related words obtained in step S52. At this time, the candidate display unit 220 also displays the similarity of the related words to the input word.

[0074] (Step S54: Word selection process) The word input unit 211 accepts the selection of a word to be used for the search from the input word and related words displayed in step S53. The word selected here is used as the input word, and the processing from step S55 onward is executed.

[0075] ***Effects of Embodiment 5*** As described above, the information management device 10 according to Embodiment 5 displays related words along with their similarity scores, allowing the user to select a word to use for the search. When the vocabulary used in the target search or how the vocabulary is used is not fully understood, referring to candidate words makes it easier to specify an appropriate word.

[0076] The embodiments and variations of this disclosure have been described above. Some of these embodiments and variations may be implemented in combination. Alternatively, some or all of them may be implemented in part. However, this disclosure is not limited to the embodiments and variations described above, and various modifications are possible as needed. [Explanation of symbols]

[0077] 10 Information management device, 11 Processor, 12 Memory, 13 Storage, 14 Communication interface, 21 Target management unit, 22 Data management unit, 23 Communication processing unit, 211 Word input unit, 212 Related word acquisition unit, 213 Search unit, 214 Evaluation value calculation unit, 215 Result output unit, 216 Similarity calculation unit, 217 Structure comparison unit, 218 Correction unit, 219 Data comparison unit, 220 Candidate display unit, 31 Management information, 32 Vocabulary list, 33 Correction rules, 34 Affiliation information.

Claims

1. The related word acquisition unit retrieves related words, which are words associated with the input word, from a vocabulary list in which multiple words representing similar objects are associated and the usage field is registered for each word. A search unit that identifies a search target by performing a search on the management information using the input word and by performing a search on the management information using the related word obtained by the related word acquisition unit, An evaluation value calculation unit calculates an evaluation value for the search target identified by the search unit, based on the similarity between the words used during the search of the search target and the input words. Equipped with, The aforementioned related word acquisition unit is an information management device that stops acquiring related words for words whose registered usage field is the same as the field to which the input word belongs.

2. The aforementioned management information includes words for each of the multiple classifications for each search target. The related word acquisition unit acquires related words for each of the input words that have been input for two or more of the multiple classifications. The search unit identifies the search target by performing a search on the management information using each combination obtained by setting the input word or related word for each of the two or more classifications, The evaluation value calculation unit calculates an evaluation value from the similarity between each word included in the combination used during the search of the search target and the input word. The information management device according to claim 1.

3. The aforementioned management information includes a knowledge graph for each search target. The evaluation value calculation unit calculates the evaluation value based on the degree of agreement between the structure of the knowledge graph for the search target identified by performing a search using the input word and the structure of the knowledge graph for the search target identified by performing a search using the related word. The information management device according to claim 1.

4. The evaluation value calculation unit calculates the evaluation value based on the similarity between the data relating to the input word for the search target identified by performing a search using the input word and the data relating to the related word for the search target identified by performing a search using the related word. The information management device according to claim 1.

5. The aforementioned information management device further, Candidate display unit that displays the related words along with the similarity between the related words and the input word. The information management device according to claim 1, comprising:

6. The aforementioned search target is an asset in any of several fields, The aforementioned vocabulary list consists of multiple words used in different fields among the aforementioned fields, and multiple words that represent similar objects are associated with each other. The information management device according to claim 1.

7. The computer retrieves related words from a vocabulary list in which multiple words representing similar objects are associated and the usage field is registered for each word, which are words associated with the input word. The computer identifies the search target by performing a search on the management information using the input word and by performing a search on the management information using the related word. The computer calculates an evaluation value for the identified search target based on the similarity between the words used during the search for the search target and the input words. An information management method in which a computer stops acquiring words as related words if the registered usage field is the same as the field to which the input word belongs.

8. The process involves retrieving related words from a vocabulary list in which multiple words representing similar objects are associated and the usage field is registered for each word, and retrieving related words that are associated with the input word. A search process to identify the search target by performing a search on the management information using the input word and performing a search on the management information using the related word obtained by the related word acquisition process, For the search target identified by the search process, an evaluation value calculation process is performed to calculate an evaluation value from the similarity between the words used during the search of the search target and the input words. The computer functions as an information management device to perform this task. The information management program in the aforementioned related word acquisition process stops acquiring words as related words if the registered usage field is the same as the field to which the input word belongs.

Images