WO2013088708A1 - 情報処理装置、情報処理システム、情報処理方法、および、コンピュータ・プログラム - Google Patents
情報処理装置、情報処理システム、情報処理方法、および、コンピュータ・プログラム Download PDFInfo
- Publication number
- WO2013088708A1 WO2013088708A1 PCT/JP2012/007930 JP2012007930W WO2013088708A1 WO 2013088708 A1 WO2013088708 A1 WO 2013088708A1 JP 2012007930 W JP2012007930 W JP 2012007930W WO 2013088708 A1 WO2013088708 A1 WO 2013088708A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- inference
- information
- inference result
- knowledge level
- knowledge
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Definitions
- the present invention relates to an information processing apparatus, an information processing system, an information processing method, and a computer program that present an inference result obtained by an inference process.
- context information is information to which an inference rule is applied.
- context information is information to which an inference rule is applied.
- character information and image information included in document data For example, character information and image information included in document data, measurement data output from sensor devices, or operation logs for devices and application software Data, etc.
- the information processing apparatus holds an inference rule expressed in an IF-THEN format as shown in the following equation (1), for example.
- the first line of Equation (1) represents a condition (IF information) that “the temperature is 100 degrees Celsius or more and the liquid is water”.
- the second line of the formula (1) represents an event (THEN information) that “the liquid changes to a gas” when the IF information is satisfied.
- the information processing apparatus using such an inference rule infers that the THEN information set for the IF information is generated when the input context information satisfies the IF information.
- such an information processing apparatus holds in advance, as inference rules, case information such as important past cases and defect cases and the conditions under which such cases may occur in the design of a certain device.
- Such an information processing apparatus extracts context information from the design document data of the apparatus, and applies an inference rule to the extracted context information.
- such an information processing apparatus can present, as an inference result, failure cases that may occur due to the design based on such design document data, important past cases related to the design, and the like.
- such an information processing apparatus can support the user's work of checking the contents of the design document data.
- Patent Document 1 accumulates past failure occurrence histories in the production process, and calculates the occurrence frequency, the degree of influence, and the degree of detection for each type of failure based on the accumulated failure occurrence history.
- An information processing device that calculates risk priority by integrating information is described.
- the information processing apparatus presents the failure information sorted by the risk priority to the user. As a result, even when there is a large amount of failure information that cannot be grasped by the user, the information processing apparatus can preferentially present failure information with a high risk priority to the user.
- Non-Patent Document 1 describes an information processing apparatus that presents information that is unknown and useful to each user by collaborative filtering using access histories of a plurality of users for a plurality of information. This information processing apparatus can present information that is unknown and useful to the user even when there is a large amount of information that cannot be grasped by the user.
- Patent Document 1 can present failure information with a high risk priority to the user, but has the following problems.
- Non-Patent Document 1 regards information that each user has not accessed as unknown information, and can present information that is determined to be useful for each user, There is a problem.
- Non-Patent Document 1 targets defect case information in design. Even if a user has already accessed defect case information related to a certain type of context information, if the type of context information is different, such a failure case information may be overlooked. In such a case, it can be said that the user does not have true detailed knowledge about the defect case information. However, since the information processing apparatus described in Non-Patent Document 1 does not present information that the user has already accessed, it may not be possible to present defect case information that does not have detailed knowledge for the user.
- information that a user has not accessed yet may be similar to information that has already been accessed. Even in this case, the information processing apparatus described in Non-Patent Document 1 may determine that information similar to information that has already been accessed is useful and present it. For such similar information, the user may already have detailed knowledge. Therefore, the information processing apparatus described in Non-Patent Document 1 may present information that the user already has detailed knowledge.
- Patent Document 1 and Non-Patent Document 1 are applied to an information processing apparatus that presents inference results, it is not possible to preferentially present inference results without detailed knowledge for each user. There is a problem that there are cases.
- the present invention has been made to solve the above-described problem, and provides an information processing apparatus that preferentially presents information that does not have detailed knowledge for each user among inference results inferred from context.
- the main purpose is to solve the above-described problem, and provides an information processing apparatus that preferentially presents information that does not have detailed knowledge for each user among inference results inferred from context. The main purpose.
- the information processing apparatus is applied until an inference result is obtained by using an inference unit that obtains an inference result by applying an inference rule to context information and information indicating a browsing user who browses the inference result.
- An inference result index value calculation unit that acquires the knowledge level of the browsing user for each inferred rule and calculates an index value that comprehensively represents the depth of knowledge of the browsing user for the inference result based on each acquired knowledge level
- an inference result presentation unit for presenting an inference result based on the index value calculated by the inference result index value calculation unit, and for each inference rule applied until the inference result is obtained, the browsing user for the inference rule
- the knowledge level is updated based on evaluation information in which the degree of knowledge of the browsing user is evaluated with respect to the inference result presented by the inference result presentation unit. It includes a level of knowledge update unit.
- the information processing system of the present invention collects context information with the above-described information processing apparatus, transmits it to the information processing apparatus, outputs an inference result presented by the information processing apparatus to the output device, and inputs it from the input device And a terminal for transmitting the evaluation information to the information processing apparatus.
- the information processing method of the present invention obtains an inference result by applying an inference rule stored in advance to the input context information, and infers each inference rule applied until the inference result is obtained.
- a value that is stored in advance as a knowledge level representing the depth of knowledge of the browsing user who browses the result, and an index that collectively represents the depth of knowledge of the browsing user with respect to the inference result based on each acquired knowledge level Calculate the value, present the inference result based on the index value, obtain evaluation information that evaluates the degree of knowledge that the browsing user has for the presented inference result, and apply each until the inference result is obtained
- the value stored as the knowledge level of the browsing user for the inference rule is updated based on the evaluation information.
- the computer program of the present invention includes a context information acquisition process for acquiring context information, an inference process for obtaining an inference result by applying an inference rule stored in advance in the storage device to the context information, an inference For each inference rule applied until the result is obtained, a value stored in advance in the storage device as a knowledge level representing the depth of knowledge possessed by the browsing user viewing the inference result is acquired, and each acquired knowledge Inference result index value calculation processing for calculating an index value that comprehensively represents the depth of knowledge of the browsing user with respect to the inference result based on the level, inference result presentation processing for presenting the inference result based on the index value, and inference Evaluation to obtain evaluation information in which the degree of knowledge held by the browsing user is evaluated for the inference result presented in the result presentation process
- Knowledge acquisition process and knowledge level update process for updating the value stored in the storage device as the knowledge level of the browsing user for the inference rule for each inference rule applied until the inference result is obtained based on the evaluation information Are executed by the computer device.
- the present invention can preferentially present information that does not have detailed knowledge for each user among the inference results inferred from the context.
- FIG. 5 is a hardware configuration diagram of a server according to second to fourth embodiments of the present invention.
- FIG. 5 is a hardware configuration diagram of a terminal according to second to fourth embodiments of the present invention.
- FIG. 1 shows a functional block configuration of the information processing apparatus 1 according to the first embodiment of the present invention.
- the information processing apparatus 1 includes an inference rule storage unit 101, a knowledge level storage unit 102, a context information acquisition unit 103, an inference unit 104, an inference result index value calculation unit 105, and an inference result presentation unit 106. And an evaluation information acquisition unit 107 and a knowledge level update unit 109.
- the hardware configuration of the information processing apparatus 1 is shown in FIG.
- the information processing apparatus 1 includes a CPU (Central Processing Unit) 2501, a RAM (Random Access Memory) 2502, a ROM (Read Only Memory) 2503, a storage device 2504 such as a hard disk, an input device 2505,
- the computer apparatus 2500 provided with the display apparatus 2506 is comprised.
- the inference rule storage unit 101 and the knowledge level storage unit 102 are configured by a storage device 2504.
- the context information acquisition unit 103 and the evaluation information acquisition unit 107 are configured by an input device 2505 and a CPU 2501 that reads a computer program stored in the ROM 2503 or the storage device 2504 into the RAM 2502 and executes it.
- the inference unit 104, the inference result index value calculation unit 105, and the knowledge level update unit 109 are configured by a CPU 2501 that reads a computer program stored in the ROM 2503 or the storage device 2504 into the RAM 2502 and executes it.
- the inference result presentation unit 106 includes a display device 2506 and a CPU 2501 that reads a computer program stored in the ROM 2503 or the storage device 2504 into the RAM 2502 and executes the computer program. Note that the hardware configuration of each functional block is not limited to the above configuration.
- the inference rule storage unit 101 stores one or more inference rules.
- Such an inference rule represents an event that occurs according to the content of context information described later.
- the inference rule may be, for example, an IF-THEN type rule including a condition (IF information) and an event (THEN information) that occurs when the condition is satisfied.
- the inference rule storage unit 101 stores such an inference rule together with an inference rule ID for identifying the inference rule.
- the knowledge level storage unit 102 stores a knowledge level representing the depth of knowledge for each user with respect to each inference rule. Specifically, the knowledge level storage unit 102 associates the inference rule ID, the user identification information (user ID), and the knowledge level indicating the depth of knowledge of the user with respect to the inference rule.
- the knowledge level storage unit 102 can be updated by a knowledge level update unit 109 described later.
- Each knowledge level stored in the knowledge level storage unit 102 may be registered in advance, or a predetermined value such as 0 may be set as an initial value.
- the context information acquisition unit 103 acquires context information.
- Context information is information to be inferred, such as character information and image information included in document data, measurement data output from sensor devices, or operation log data for devices and application software, etc. Also good.
- the inference unit 104 obtains an inference result by applying the inference rules stored in the inference rule storage unit 101 to the acquired context information. For example, the inference unit 104 searches for inference rules having IF information that matches the context information. Then, the inference unit 104 estimates that the THEN information included in the retrieved inference rule occurs. In addition, when the THEN information of the applied inference rule satisfies the IF information of another inference rule, the inference unit 104 may repeat a chain inference process that further applies the other inference rule. If there are a plurality of inference rules applicable to the acquired context information or the THEN information of the applied inference rule, the inference unit 104 may execute a plurality of inference processes based on the respective inference rules.
- the inference result index value calculation unit 105 calculates an index value of the inference result by using the browsing user's knowledge level for each inference rule applied until the inference result is obtained.
- the browsing user is a user who browses the inference result.
- the index value of the inference result is a value that comprehensively represents the depth of knowledge of the browsing user with respect to the inference result.
- the inference result index value calculation unit 105 may acquire the user ID of the viewing user from the inference result presentation unit 106 described later. Then, the inference result index value calculation unit 105 may acquire the knowledge level associated with each inference rule ID and the browsing user ID from the knowledge level storage unit 102. Then, the inference result index value calculation unit 105 may calculate an inference result index value for the browsing user based on each acquired knowledge level. For example, the inference result index value calculation unit 105 may calculate the acquired integrated value of each knowledge level as an index value. As described above, if the value of the knowledge level is smaller, the inference derived using the inference rule that the browsing user does not have detailed knowledge if the viewing user indicates that the inspecting rule is not detailed.
- the index value as the integrated value becomes smaller.
- an inference result derived by an inference process in which the viewing user does not have detailed knowledge is considered to be more useful for the viewing user. Therefore, in this case, the index value for the inference result indicates that the smaller the value, the higher the degree of usefulness for the user.
- the inference result index value calculation unit 105 calculates an index value for the viewing user for each inference result.
- the inference result presentation unit 106 presents an inference result based on the index value calculated by the inference result index value calculation unit 105. For example, the inference result presentation unit 106 may determine the order in which the inference results are presented based on the index value. If the index value is smaller, the inference result presenting unit 106 may present the inference results in ascending order of the index value if the degree of usefulness for the browsing user is higher.
- the inference result presentation unit 106 acquires the user ID of the browsing user who browses the inference result, and notifies the inference result index value calculation unit 105 of the user ID.
- the inference result presentation unit 106 may display an input screen for a browsing user ID and present an inference result when a registered browsing user ID is input.
- the evaluation information acquisition unit 107 acquires evaluation information obtained by evaluating the degree of knowledge of the browsing user with respect to the inference result presented by the inference result presentation unit 106. For example, the evaluation information acquisition unit 107 may acquire evaluation information indicating whether or not the inference result is known to the browsing user via the input device.
- the knowledge level update unit 109 uses the evaluation information acquired by the evaluation information acquisition unit 107 as the value stored in the knowledge level storage unit 102 as the browsing user's knowledge level for each inference rule applied until the inference result is obtained. Update based on. For example, when the evaluation information indicating whether or not the evaluation information acquisition unit 107 is known, the knowledge level update unit 109 determines the knowledge level as long as the evaluation information indicates “known”. Values may be added. Further, if the evaluation information represents “not known”, the knowledge level update unit 109 may subtract a predetermined value from the knowledge level.
- FIG. 2 is a flowchart for explaining the inference result presentation operation of the information processing apparatus according to the first embodiment of the present invention.
- a set of two hexagons indicates that a series of processes sandwiched therebetween is repeated (loop).
- the hexagon which is a quadrangular shape with the upper corners cut off, represents the beginning of repeated processing.
- the hexagon which is a square shape with the lower corners cut off represents the end of the repeated processing. It should be noted that the object of the repetition process may be described in a hexagon representing the start of repetition.
- the context information acquisition unit 103 acquires context information (step S1).
- the context information acquisition unit 103 may extract context information from given document data.
- the inference unit 104 obtains an inference result by applying the inference rule stored in the inference rule storage unit 101 to the context information acquired in step S1 (step S2). At this time, as described above, the inference unit 104 may acquire an inference result by executing inference processing in a chained manner.
- the inference result index value calculation unit 105 acquires a browsing user ID for browsing the inference result (step S3). As described above, the inference result index value calculation unit 105 may acquire the browsing user ID from the inference result presentation unit 106.
- the inference result index value calculation unit 105 executes the processing of steps S4 to S5 for each inference result obtained in step S2.
- the inference result index value calculation unit 105 searches the knowledge level storage unit 102 for the knowledge level of the viewing user for each inference rule used by the inference unit 104 until the inference result is obtained (step S4).
- the inference result index value calculation unit 105 calculates an index value that comprehensively represents the depth of knowledge of the browsing user with respect to the inference result based on the knowledge level searched in step S4 (step S5). For example, as described above, the inference result index value calculation unit 105 may calculate an integrated value of the browsing user's knowledge level for each inference rule used until the inference result is obtained as an index value.
- the inference result presentation unit 106 presents an inference result based on the index value calculated by the inference result index value calculation unit 105 (step S6).
- the inference result presentation unit 106 may present the inference results in ascending order of the index value.
- the information processing apparatus 1 ends the inference result presentation operation.
- FIG. 3 is a flowchart for explaining the knowledge level update operation of the information processing apparatus according to the first embodiment of the present invention.
- the evaluation information acquisition unit 107 acquires the evaluation information of the viewing user for one of the inference results presented in step S6 of FIG. 2 (step S11). For example, as described above, the evaluation information acquisition unit 107 inputs the evaluation information indicating whether or not the inference result is known by displaying the inference result on the display device in a selectable manner. You may make it acquire via an apparatus.
- the knowledge level update unit 109 executes the following steps S12 to S13 for each inference rule used in step S2 in FIG. 2 until this inference result is obtained.
- the knowledge level update unit 109 searches the knowledge level storage unit 102 for the knowledge level associated with the inference rule ID and the browsing user ID (step S12).
- the knowledge level update unit 109 updates the knowledge level of the retrieved record based on the type of evaluation information acquired in step S11 (step S13). For example, the knowledge level update unit 109 may add or subtract a predetermined value for the knowledge level of the corresponding record in accordance with the evaluation information acquired in step S11.
- the information processing apparatus 1 ends the knowledge level update operation. Note that when the information processing apparatus 1 presents a plurality of inference results, the information processing apparatus 1 may perform such a knowledge level update operation on each inference result.
- the information processing apparatus 1 may not include some or all of the inference rule storage unit 101, the knowledge level storage unit 102, the context information acquisition unit 103, and the evaluation information acquisition unit 107. Further, these components may be outside the information processing apparatus 1. Furthermore, some or all of the inference rules, knowledge level, context information, and evaluation information may be given to the information processing apparatus 1 from the outside.
- the information processing apparatus can preferentially present information that the user does not have detailed knowledge among the information inferred from the context.
- the knowledge level storage unit stores the browsing user's knowledge level for the inference rule
- the inference result index value calculation unit is based on the browsing user's knowledge level for each inference rule used in the inference process.
- an index value that comprehensively represents the depth of knowledge of the browsing user with respect to the inference result is calculated, and the inference result presentation unit presents the inference result based on the calculated index value.
- the information processing apparatus according to the first embodiment determines and presents the usefulness of the inference result for the viewing user based on the depth of knowledge of the viewing user with respect to the inference process.
- the information processing apparatus according to the first embodiment can preferentially present an inference result that has a low knowledge level for the inference process, even if the inference result is known to the browsing user. is there.
- the evaluation information acquisition unit acquires the browsing user evaluation information for the presented inference result
- the knowledge level update unit browses the browsing user's knowledge level for each inference rule used in the inference process. This is because the update is performed based on the user evaluation information.
- the information processing apparatus can update the browsing user's knowledge level for each inference rule used in the inference process based on the evaluation of the browsing user's knowledge level with respect to the inference result. it can. For example, if evaluation information indicating that the presented inference result is known is obtained, the information processing apparatus according to the first embodiment allows the viewing user's knowledge level for each inference rule until the inference result is obtained. Can be updated higher than before.
- the information processing apparatus updates the knowledge level of the browsing user for each of these inference rules to a lower level than before. can do.
- the information processing apparatus can calculate an index value for other inference results using such an inference rule in the inference process with higher accuracy.
- the information processing system 20 includes a server 2 as an information processing apparatus of the present invention and a terminal 8.
- the functional block configuration of each device is shown in FIG.
- the server 2 includes an inference rule storage unit 201, a knowledge level storage unit 202, a context information acquisition unit 203, an inference unit 204, an inference result index value calculation unit 205, an inference result presentation unit 206, An evaluation information acquisition unit 207, a knowledge level change rule storage unit 208, a knowledge level update unit 209, and a case information storage unit 210 are provided.
- the server 2 includes a computer device 2600 that includes a CPU 2601, a RAM 2602, a ROM 2603, a storage device 2604, and a network interface 2605.
- the inference rule storage unit 201, the knowledge level storage unit 202, the knowledge level change rule storage unit 208, and the case information storage unit 210 are configured by a storage device 2604.
- the context information acquisition unit 203, the evaluation information acquisition unit 207, and the inference result presentation unit 206 are a network interface 2605, and a CPU 2601 that reads a computer program stored in the ROM 2603 or the storage device 2604 into the RAM 2602 and executes it.
- the inference unit 204, the inference result index value calculation unit 205, and the knowledge level update unit 209 are configured by a CPU 2601 that reads a computer program stored in the ROM 2603 or the storage device 2604 into the RAM 2602 and executes it.
- the terminal 8 includes a context information collection unit 801 and an information input / output unit 802.
- the terminal 8 includes a computer device 2700 including a CPU 2701, a RAM 2702, a ROM 2703, a storage device 2704, an input device 2705, a display device 2706, and a network interface 2707.
- the context information collection unit 801 includes a network interface 2707 and a CPU 2701 that reads a computer program stored in the ROM 2703 or the storage device 2704 into the RAM 2702 and executes the computer program.
- the information input / output unit 802 includes an input device 2705, a display device 2706, a network interface 2707, and a CPU 2701 that reads a computer program stored in the ROM 2703 or the storage device 2704 into the RAM 2702 and executes it.
- each functional block of each device is not limited to the above-described configuration.
- the server 2 and the terminal 8 are communicably connected via a network constituted by the Internet, a LAN (Local Area Network), a public line network, a wireless communication network, or a combination thereof.
- 4 shows one terminal 8, the number of terminals to which the information processing apparatus of the present invention is connected is not limited.
- the case information storage unit 210 stores case information representing cases related to inference rules.
- An example of information stored in the case information storage unit 210 is shown in FIG. In FIG. 5, the case information represented by each line includes information (case ID) for identifying the case, the name of the case, and a related URL (UniformUniResource Locator).
- the case information may include the registration date and the user ID of the registrant.
- the inference rule storage unit 201 stores each inference rule described in the first embodiment of the present invention further including a case ID.
- An example of information stored in the inference rule storage unit 201 is shown in FIG.
- the inference rule indicated by each row includes an inference rule ID, IF information, THEN information, and a case ID.
- the inference rule may include the registration date and the user ID of the registrant.
- the inference rule may include a plurality of IF information.
- an inference rule including a plurality of IF information means that THEN information is inferred when all IF information is satisfied.
- an inference rule including a plurality of IF information may mean that THEN information is inferred when any of the plurality of IF information is satisfied.
- the inference rule including a plurality of IF information may further include information indicating either “and condition” or “or condition” as an application condition of the own rule.
- the “and” condition means that it is applied when all IF information is satisfied.
- the “or condition” means that it is applied when any IF information is satisfied.
- the inference rule may include a plurality of THEN information.
- the case ID included in the inference rule indicates case information related to the inference rule. If there is no case information related to a certain inference rule, the inference rule storage unit 201 may not include the case ID in the inference rule.
- the knowledge level storage unit 202 stores a value included in a predetermined range as a knowledge level for each user for each inference rule.
- An example of information stored in the knowledge level storage unit 202 is shown in FIG.
- the range that the knowledge level can take is assumed to be a range from 0 to 1.
- the smaller the value the less detailed the user is about the inference rule.
- the knowledge level of the user U-0001 for the inference rule P-0001 is 0.6
- the knowledge level for the inference rule P-0004 is 0.1.
- the knowledge of the inference rule P-0004 of the user U-0001 is less detailed than the knowledge of the inference P-0001.
- the range that the knowledge level can take is not limited to this and can be set freely.
- the context information acquisition unit 203 receives context information from the terminal 8 via the network.
- the inference unit 204 obtains case information as an inference result by sequentially applying the inference rules stored in the inference rule storage unit 201 to the context information acquired by the context information acquisition unit 203.
- applying in a chain means that another inference rule having IF information that is satisfied by the THEN information of the applied inference rule is further applied.
- the inference unit 204 ends the inference when there are no more applicable inference rules. If a plurality of inference rules are applicable in the inference process, the inference unit 204 continues the inference processing in a chain manner for each of the plurality of inference rules.
- the inference unit 204 obtains the case information indicated by the case ID associated with each inference rule applied in the inference process as the inference result. That is, the inference unit 204 obtains case information related to each inference rule applied in the inference process as an inference result. In addition, the inference unit 204 assumes that no inference result is obtained if there is no case ID associated with each inference rule applied until the inference is completed.
- a list of inference rules until an inference rule related to the case information is applied is called an inference step list of the case information.
- the inference result index value calculation unit 205 calculates an index value as a browsing user for each case information derived as an inference result based on the inference step list of the case information. For example, the inference result index value calculation unit 205 may calculate an index value for the browsing user of the case information obtained as the inference result using the following equation (2).
- a represents the user ID
- L (a, k) represents the knowledge level of the user a with respect to the inference rule of the inference step k.
- Case represents case information as an inference result
- Root represents an inference step list of the case information Case.
- k represents the application order of each inference rule (inference step) included in the inference step list Root
- S (a, Case) represents an index value of the case Case for the user a.
- one case information Case may be obtained by a plurality of inference step lists Root. Therefore, the expression (2) selects the one with the smallest accumulated value of the knowledge level of the browsing user for each included inference rule from one or more inference step lists Root until the case information Case is obtained.
- the index value for the user a of the case information Case is represented.
- the inference result index value calculation unit 205 may use an average value or a power average of knowledge level integrated values calculated in a plurality of inference step lists as an index value for a case information browsing user. In addition, the inference result index value calculation unit 205 may calculate not only the integrated value of the browsing user's knowledge level for each inference rule included in the inference step list but also the sum, minimum value, sum of powers, and the like. .
- the inference result presentation unit 206 sorts or filters the case information as the inference result obtained by the inference unit 204 based on the index value, and transmits the result to the terminal 8. For example, when the index value having a lower value indicates that the usefulness is higher, the inference result presentation unit 206 may transmit the case information sorted in ascending order of the index value to the terminal 8. In addition, the inference result presentation unit 206 may transmit, for example, case information whose index value is equal to or less than a threshold value to the terminal 8.
- the evaluation information acquisition unit 207 receives from the terminal 8 one of a plurality of predetermined types as evaluation information in which the degree of knowledge of the browsing user with respect to the case information as the inference result is evaluated. For example, the evaluation information acquisition unit 207 notifies the terminal 8 to display a plurality of predetermined types of evaluation information in a selectable display format such as a drop-down list for each case information as an inference result. May be. In this case, the evaluation information acquisition unit 207 receives the type of evaluation information selected by the user operation on the terminal 8 from the terminal 8.
- the knowledge level change rule storage unit 208 stores a knowledge level change rule in which the type of evaluation information is associated with the increase / decrease value of the knowledge level.
- An example of information stored in the knowledge level change rule storage unit 208 is shown in FIG. In FIG. 8, the knowledge level change rule represented by each row includes information representing the type of evaluation information and an increase / decrease value of the knowledge level.
- evaluation information of the type “specialized area” is associated with information “+0.5”, which means that a value obtained by adding 0.5 to the previous knowledge level is used as a new knowledge level. .
- “Browse” indicates that the browse user has accessed the URL associated with the case information that is the inference result. That is, for the case information as the inference result, the browsing user represents an evaluation that the knowledge is deepened by browsing. Therefore, the knowledge level change rule for this evaluation information is assumed that the knowledge level of the browsing user for each inference rule applied until the case information evaluated as “browsing” is deepened to some extent. This means that 0.1 is added.
- “Verification execution” Indicates that the browsing user has actually verified that such case information is guided by such an inference basis. That is, for the case information as the inference result, the browsing user represents an evaluation that the knowledge is deepened by performing the verification.
- the knowledge level change rule for this evaluation information is based on the assumption that the browsing user's knowledge for each inference rule applied until the case information evaluated as “verification performed” is deepened to some extent. It represents adding 0.2.
- “Specialized area” indicates that the browsing user already has specialized knowledge about the inference result and the reasoning reason. That is, the browsing user represents an evaluation that the user already has detailed knowledge with respect to the case information as the inference result. Therefore, the knowledge level change rule for this evaluation information is based on the assumption that the browsing user's knowledge for each inference rule applied until the case information evaluated as “specialized area” was originally deep. It represents adding 0.5 to the level.
- “Overlooked” This indicates that the browsing user has overlooked the inference result based on such an inference basis and noticed only after the inference result is presented.
- the browsing user represents an evaluation that the case information as the inference result is overlooked so that the browsing user does not have detailed knowledge about the case. Therefore, the knowledge level change rule for this evaluation information is assumed to be 0. 0 from each knowledge level, assuming that the browsing user's knowledge for each inference rule applied until the case information evaluated as “occurrence of oversight” is derived is shallow. This means that 5 is subtracted.
- “Defect Embedding” This indicates that the browsing user is an implementer of the defect case or actually caused the same defect as the defect case.
- the browsing user represents an evaluation that the user does not have detailed knowledge about the case as the inference result is actually implemented. Therefore, the knowledge level change rule for this evaluation information is based on the assumption that the browsing user's knowledge for each inference rule applied until the case information evaluated as “defect embedding” is very shallow, This means that 1.0 is subtracted.
- the knowledge level update unit 209 acquires an increase / decrease value associated with the type of evaluation information acquired by the evaluation information acquisition unit 207 from the knowledge level change rule storage unit 208. Further, the knowledge level update unit 209 searches the knowledge level storage unit 202 for a record that stores the knowledge level of the browsing user for each inference rule included in the inference step list for the case information that is the target of the evaluation information. And the knowledge level update part 209 updates the knowledge level of each searched record according to the acquired knowledge level change rule.
- the knowledge level update unit 209 updates the knowledge level of the target record stored in the knowledge level update unit 209 within the range of 0 to 1. For example, the knowledge level update unit 209 updates the knowledge level exceeding 1 to the upper limit of 1 by adding according to the knowledge level change rule. Also, the knowledge level update unit 209 updates the knowledge level that becomes negative by subtraction according to the knowledge level change rule to 0, which is the lower limit value. Note that the knowledge level update unit 209 is not limited to the range of 0 to 1, but may update the knowledge level within a predetermined range.
- the context information collection unit 801 collects context information from the target data.
- the context information collection unit 801 may collect information representing each component and its attribute information as context information from design document data stored in the storage device. Then, the context information collection unit 801 transmits the collected context information to the server 2.
- the information input / output unit 802 displays the inference result received from the server 2 on the display device.
- the received inference result is case information sorted or filtered based on the index value as described above.
- the information input / output unit 802 acquires evaluation information for the displayed inference result via the input device. At this time, the information input / output unit 802 displays a plurality of types of evaluation information notified from the server 2 on a display device in a selectable form such as a drop-down list, so that any type of evaluation information is displayed. You may get it. Then, the information input / output unit 802 transmits the acquired type of evaluation information to the server 2.
- FIG. 9 is a flowchart for explaining the inference result presentation operation of the information processing system according to the second embodiment of the present invention.
- the left diagram represents the operation of the terminal 8
- the right diagram represents the operation of the server 2
- the dashed arrows connecting the left and right represent the data flow.
- the context information collection unit 801 of the terminal 8 collects context information and transmits it to the server 2 (step S20).
- the context information acquisition unit 203 of the server 2 receives the context information (step S21).
- the inference unit 204 applies the inference rules stored in the inference rule storage unit 201 in a chain manner to the received context information. Then, when there are no more applicable inference rules, the inference unit 204 acquires a case ID associated with each applied inference rule as an inference result (step S22).
- the inference result index value calculation unit 205 acquires a browsing user ID (step S23).
- the inference result index value calculation unit 205 executes the processing of steps S24 to S25 for each case ID obtained in step S22.
- the inference result index value calculation unit 205 searches the knowledge level storage unit 202 for the knowledge level of the browsing user for each inference rule included in the inference step list of this case information (step S24).
- the inference result index value calculation unit 205 calculates the index value of this case information by applying the knowledge level searched in step S24 to equation (2) (step S25).
- the inference result presentation unit 206 notifies the terminal 8 to present the case information indicated by each case ID obtained in step S22 based on the index value calculated in step S25 (step S26). For example, the inference result presentation unit 206 may notify the terminal 8 to sort and present case information in ascending order of index values. Further, the inference result presenting unit 206 may perform filtering on the condition whether or not the index value is equal to or less than the threshold value, and notify the terminal 8 to present the case information based on the result.
- the information input / output unit 802 of the terminal 8 presents case information as an inference result in accordance with the notification from the server 2 (step S27).
- the information processing system 20 ends the inference result presentation operation.
- FIG. 10 is a flowchart for explaining the knowledge level update operation of the information processing system according to the second embodiment of the present invention.
- the left diagram represents the operation of the terminal 8
- the right diagram represents the operation of the server 2
- the dashed arrows connecting the left and right represent the data flow.
- the information input / output unit 802 of the terminal 8 displays a plurality of types of evaluation information so as to be selectable as the degree of knowledge of the browsing user with respect to the case information as the inference result (step S30).
- the plurality of types of evaluation information is notified from the server 2 in advance.
- the plurality of types of evaluation information may be information indicating “browsing”, “verification performed”, “specialized area”, “occurrence of oversight”, and “defect embedding”, as described above.
- the information input / output unit 802 transmits the type of evaluation information selected by the operation of the input device for one of the presented case information to the server 2 (step S31).
- the evaluation information acquisition unit 207 of the server 2 receives the type of evaluation information from the terminal 8 (step S32).
- the knowledge level update unit 209 searches the knowledge level change rule storage unit 208 for knowledge level change rules regarding the type of evaluation information acquired in step S32 (step S33).
- the knowledge level update unit 209 executes the processing of steps S34 to S35 for each inference rule included in the inference step list of the target case information.
- the knowledge level updating unit 209 searches the knowledge level storage unit 202 for a record including the browsing user's knowledge level for this inference rule (step S34).
- the knowledge level update unit 209 updates the knowledge level of the retrieved record within a predetermined range according to the knowledge level change rule acquired in step S33 (step S35).
- the information processing system 20 ends the knowledge level update operation.
- FIG. 11 is a schematic diagram for explaining a specific example of the inference result presentation operation of the information processing system according to the second embodiment of the present invention.
- the circuit design drawing data is stored in the terminal 8.
- the context information collection unit 801 of the terminal 8 extracts part information and attribute information of each part as context information from the circuit design drawing data.
- the context information collection unit 801 extracts context information “case Y” as component information and “plastic”, “sliding”, and “high temperature” as attribute information from the circuit design drawing data ( Step S20).
- the context information acquisition unit 203 of the server 2 receives this context information from the terminal 8 (step S21).
- the inference unit 204 determines that the attribute information “plastic” and “high temperature” of the context information satisfy the IF information of the inference rule P-0001 shown in FIG.
- the inference unit 204 determines that the THEN information “part ⁇ melt” of the inference rule P-0001 and the attribute information “plastic” of the case Y satisfy the IF information of the inference rule P-0004.
- the inference unit 204 ends the inference process because there is no inference rule in which the THEN information “part ⁇ melt deformation” of the inference rule P-0004 satisfies the IF information.
- the inference unit 204 acquires a case ID associated with each of the inference rules P-0001 and P-0004 used in step S22.
- no case ID is associated with P-0001
- C-0001 is associated with P-0004 as the case ID. Therefore, the inference unit 204 acquires a case C-0001 as an inference result (step S22).
- the inference unit 204 applies other inference rules (not shown) to the attribute information “slide type” of the context information obtained in step S21 in a chained manner as shown in FIG. Cases C-0007 to C-0009 are obtained as inference results.
- case ID is not associated with the inference rule applied during the inference process. If there is a case ID associated with an inference rule applied midway, the inference unit 204 also uses the case ID as an inference result.
- the inference result index value calculation unit 205 calculates an index value for each browsing ID for the browsing user.
- the browsing user ID is U-0001.
- the inference result index value calculation unit 205 obtains ⁇ inference rule P-0001, inference rule P-0004 ⁇ as an inference step list for the case C-0001 as the inference result.
- the inference result index value calculation unit 205 obtains 0.6 by referring to the knowledge level storage unit 202 shown in FIG. 7 as the knowledge level of the user U-0001 for the inference rule P-0001 (step S24). ).
- the inference result index value calculation unit 205 obtains 0.1 as a knowledge level of the user U-0001 for the inference rule P-0004 by referring to the knowledge level storage unit 202 shown in FIG. S24).
- the inference result index value calculation unit 205 obtains 0.06 as an index value for the user U-0001 of the case C-0001 using the formula (2) (step S25).
- the inference result index value calculation unit 205 calculates 0.5, 0.3, and 0.08 as knowledge levels for cases C-0007 to C-0009, respectively.
- the inference result index value calculation unit 205 determines each of the inference result index value calculation units 205 based on the equation (2). You may employ
- the inference result presentation unit 206 searches the case information storage unit 210 for case information of cases C-0001 and C-0007 to C-0009, which are inference results. Then, the inference result presenting unit 206 sorts each case information in ascending order of the index values calculated in Step S25, and transmits it to the terminal 8 (Step S26).
- FIG. 12 is a diagram showing an example of an inference result presentation screen according to the second embodiment of the present invention.
- the information input / output unit 802 of the terminal 8 displays the case information that is the inference result in order from the lowest index value.
- each line has shown the case information which is an inference result.
- the “uninput” operation button shown in the rightmost cell of each row is for opening an information input screen for evaluating the degree of knowledge of the browsing user with respect to the case information.
- “input completed” information shown instead of the “uninput” operation button indicates that the evaluation information on the degree of knowledge of the browsing user with respect to this case information has already been input.
- FIG. 13 is an example of an evaluation information input screen displayed by the information input / output unit 802 of the terminal 8.
- This example is a screen for inputting evaluation information for case C-0001.
- the information input / output unit 802 displays the input screen illustrated in FIG. 13 in response to pressing of the “uninput” operation button illustrated in FIG. 12.
- the information input / output unit 802 can select any one of “browsing”, “verification performed”, “specialized area”, “occurrence of oversight”, and “defect embedding” as multiple types of evaluation information.
- a pull-down list is displayed (step S30).
- the information input / output unit 802 acquires evaluation information “verification performed” (steps S31 to S32).
- FIG. 14 is a schematic diagram for explaining a specific example of the knowledge level update operation of the information processing system according to the second embodiment of the present invention.
- the knowledge level update unit 209 acquires P-0001 and P-0004 as the IDs of the inference rules included in the inference step list until the case C-0001 is obtained. Then, the knowledge level update unit 209 searches the knowledge level storage unit 202 for records including P-0001 and the browsing user U-0001, and records including P-0004 and the browsing user U-0001 (step S34). .
- the knowledge level updating unit 209 updates the knowledge level included in the searched record to a value obtained by adding 0.2 to the value.
- the knowledge level update unit 209 uses the knowledge level of the user ID for the corresponding inference rule ID as 0, and uses the value obtained by applying the increase / decrease value as a new record. You may store in the level memory
- the server 2 stores some or all of the inference rule storage unit 201, the knowledge level storage unit 202, the context information acquisition unit 203, the evaluation information acquisition unit 207, the knowledge level change rule storage unit 208, and the case information storage unit 210. It does not have to be provided. Further, these components may be outside the server 2. Furthermore, some or all of inference rules, knowledge levels, context information, evaluation information, knowledge level change rules, and case information may be given to the server 2 from the outside.
- the information processing system can preferentially present the case information that the user does not have detailed knowledge among the case information inferred from the context.
- the inference section uses the case information associated with each inference rule that is applied to the context in a chain as the inference result, and the inference result evaluation section uses the inference step list until the case information is obtained.
- an index value that comprehensively represents the depth of knowledge of the browsing user for the case information is calculated.
- the information processing system according to the second embodiment provides such case information to a user who knows the case information but does not have detailed knowledge of the process up to the case. It can be preferentially presented. That is, the information processing system according to the second embodiment can preferentially present the case information as the inference result from the less knowledge the browsing user has for the inference process.
- the information processing system according to the second embodiment is able to accurately present case information that can be said to be case information that cannot be said to have truly detailed knowledge for the user, and that is likely to make a mistake. it can.
- the information processing system can more accurately determine the case information presented with priority for each browsing user.
- the evaluation information acquisition unit acquires information in which the degree of knowledge of the case information as the inference result is evaluated by the browsing user, and the knowledge level update unit changes the knowledge level associated with the acquired evaluation information.
- the knowledge level of the viewing user for each inference rule used until the case information as the inference result is obtained using the rule is updated.
- the information processing system according to the second embodiment improves the accuracy of index values of other inference results obtained by using the inference rule whose knowledge level has been updated.
- the information processing system according to the second embodiment even if the case information has never been viewed by the user, if the user already has detailed knowledge about the inference process, Case information is not preferentially displayed to the user.
- the information processing system is likely to cause mistakes even if it is case information that has been viewed many times by the user, because it does not have true detailed knowledge about the inference process. If it is information, the case information can be preferentially displayed to the user.
- the information processing system 30 includes a server 3 as an information processing apparatus of the present invention and a terminal 8.
- the functional block configuration of each device is shown in FIG. 15, the server 3 replaces the inference rule storage unit 201 with the inference rule storage unit 301 and the knowledge level storage unit 202 with respect to the server 2 according to the second embodiment of the present invention.
- An inference result index value calculation unit 305 instead of the storage unit 302, an inference result index value calculation unit 205, a knowledge level update unit 309 instead of the knowledge level update unit 209, and a case information storage instead of the case information storage unit 210
- the point provided with the part 310 differs.
- the server 3 is configured by a computer device 2600 shown in FIG.
- the inference rule storage unit 301, the knowledge level storage unit 302, the inference result index value calculation unit 305, the knowledge level update unit 309, and the case information storage unit 310 are the inference rule storage unit 201, the knowledge level storage unit 202, and the inference result index.
- the value calculation unit 205, the knowledge level update unit 209, and the case information storage unit 210 are configured by the same components of the computer device 2600.
- the inference rule storage unit 301 stores each inference rule as exemplified in FIG. 6 in the second embodiment of the present invention, further including its occurrence probability.
- An example of information stored in the inference rule storage unit 301 is shown in FIG. In FIG. 16, for example, the inference rule P-0002 includes “0.5” as the occurrence probability. This indicates that when the condition that the output from the regulator component is a high voltage is satisfied, the probability that the component generates heat is 0.5.
- Knowledge level storage unit 302 further stores records including case IDs instead of inference rule IDs in addition to the records illustrated in FIG. 7 in the second embodiment of the present invention.
- An example of information stored in the knowledge level storage unit 302 is shown in FIG. In FIG. 17, each record stored in the knowledge level storage unit 302 includes one of an inference rule ID and a case ID, a user ID, and a knowledge level.
- the record including the inference rule ID is the same as the record in the first and second embodiments of the present invention, and represents the knowledge level of the user with respect to the inference rule.
- the record including the case ID represents the user's knowledge level for the case information.
- the user's knowledge level for such case information may be registered in advance, or a predetermined value such as 0 may be set as an initial value.
- the case information storage unit 310 stores information representing the importance of the case in addition to the configuration of the case information in the second embodiment of the present invention as information representing the case related to the inference result. Yes.
- An example of information stored in the case information storage unit 310 is shown in FIG.
- the inference result index value calculation unit 305 for each piece of case information derived as an inference result, in addition to the viewing user's knowledge level for each inference rule included in the inference step list, further provides the viewing user's knowledge for the case information. Based on the level, the index value of the case information is calculated. For example, the inference result index value calculation unit 305 may calculate the index value of each case information as the inference result using the following equation (3).
- Lc (a, Case) is a knowledge level for the case Case of the user a.
- the inference result index value calculation unit 305 may calculate an index value for the index value of each case information calculated as described above in consideration of the importance of the case information. For example, the inference result index value calculation unit 305 may set a new index value by adding the reciprocal of the importance of the case information to the index value calculated using the formula (3). Thereby, the index value of the case information with higher importance of the case information itself becomes small among the case information whose browsing user's knowledge is lower than the inference process of the case information.
- the inference result index value calculation unit 305 may calculate the index value of the case information in consideration of the occurrence probability of the event for each inference rule as shown in FIG. For example, the inference result index value calculation unit 305 adds a value obtained by adding the power of each occurrence probability to the integrated value of the browsing user's knowledge level for each inference rule included in the inference step list of the case information. It may be an index value. If a positive exponent is used to multiply the occurrence probability power, the index value calculated for the case information obtained through the inference rule with the lower occurrence probability is smaller.
- the inference result index value calculation unit 305 regards case information as an inference result inferred from an event with a lower occurrence frequency as being more useful (less knowledge) for the viewing user.
- An index value is calculated.
- Such an index value is effective, for example, when the browsing user is an advanced user. Or, when the power of the occurrence probability is integrated using a negative exponent, the index value calculated for the case information obtained through the inference rule with the higher occurrence probability becomes smaller.
- the inference result index value calculation unit 305 regards the case information as an inference result inferred from an event having a higher occurrence frequency as an index that is considered more useful (less knowledge) for the viewing user. Calculate the value.
- Such an index value is effective, for example, when the browsing user is a beginner.
- the inference result index value calculation unit 305 may change the index value in the power of occurrence probability used when calculating the index value by acquiring information indicating the skill level of the browsing user.
- Knowledge level update unit 309 is configured in the same manner as knowledge level update unit 209 in the second embodiment of the present invention.
- the knowledge level update unit 309 searches the knowledge level storage unit 302 for a record including the browsing user's knowledge level for the case information that is the target of the evaluation information. Then, the knowledge level update unit 309 updates the knowledge level of the retrieved record according to a knowledge level change rule corresponding to the type of evaluation information. That is, the knowledge level update unit 309 updates the browsing user's knowledge level for the case information that is the target of the evaluation information based on the evaluation information.
- FIG. 19 is a flowchart for explaining the inference result presentation operation of the information processing system according to the third embodiment of the present invention.
- the information processing system 30 operates in the same manner as the information processing system 20 according to the second embodiment of the present invention from step S20 to S23, thereby chaining inference rules with respect to context information. And apply case ID as an inference result.
- the inference result index value calculation unit 305 executes the following steps S41 to S45 for each case ID obtained in step S22.
- the inference result index value calculation unit 305 acquires the browsing user's knowledge level for each inference rule included in the inference step list until the case information is obtained (step S41).
- the inference result index value calculation unit 305 acquires the browsing user's knowledge level for the case information (step S42).
- the inference result index value calculation unit 305 acquires the probability of occurrence of each inference rule included in the inference step list until the case information is obtained (step S43).
- the inference result index value calculation unit 305 acquires the importance of the case information (step S44).
- the inference result index value calculation unit 305 browses this case information based on each knowledge level acquired in steps S41 to S42, each occurrence probability acquired in step S43, and the importance acquired in step S44. An index value for the user is calculated (step S45).
- the inference result index value calculation unit 305 performs steps S41 to S41 for each inference step list. S45 is executed. Then, the inference result index value calculation unit 305 determines the index value of the case information based on the obtained plurality of index value candidates. For example, the inference result index value calculation unit 305 may use the minimum value of a plurality of index values as the index value of this case ID.
- the information processing system 30 When the index value calculation processing is completed for each case ID, the information processing system 30 operates in the same manner as in the second embodiment of the present invention from step S26 to S27, and presents case information based on the index value. .
- the information processing system 30 ends the inference result presentation operation.
- FIG. 20 is a flowchart for explaining the knowledge level update operation of the information processing system according to the third embodiment of the present invention.
- the information processing system 30 operates in the same manner as the information processing system 30 according to the second embodiment of the present invention from step S30 to S35, and thus each inference until the case information as an inference result is obtained. Update the browsing user's knowledge level for the rule.
- the knowledge level update unit 309 searches the knowledge level storage unit 302 for a record including the browsing user's knowledge level for the case information that is the target of the evaluation information acquired in step S32 (step S51).
- the knowledge level update unit 309 updates the knowledge level of the record searched in step S51 according to the knowledge level change rule obtained in step S33 (step S52).
- the information processing system 30 ends the knowledge level update operation.
- the server 3 stores some or all of the inference rule storage unit 301, the knowledge level storage unit 302, the context information acquisition unit 203, the evaluation information acquisition unit 207, the knowledge level change rule storage unit 208, and the case information storage unit 310. It does not have to be provided. Further, these components may be outside the server 3. Furthermore, some or all of inference rules, knowledge levels, context information, evaluation information, knowledge level change rules, and case information may be given to the server 3 from the outside.
- the information processing apparatus more accurately determines case information that the user does not have detailed knowledge as case information preferentially presented from case information inferred from the context. can do.
- the inference result index value calculation unit in addition to the viewing user's knowledge level for each inference rule included in the inference step list from which the case information as the inference result was derived, This is because the index value of the case information is calculated in consideration of the above.
- the inference result index value calculation unit takes into account the probability of occurrence of each inference rule included in the inference step list from which the case information as the inference result is derived, the importance of the case information itself, and the like. It is because it calculates.
- the information processing system is not limited to the user's knowledge level for the inference process, but also the probability of occurrence of each inference process, the user's knowledge level for the case itself, the importance of the case itself, Alternatively, by considering these combinations, it is possible to determine preferentially presented case information based on a more accurate index value.
- An information processing system 40 includes a server 4 and a terminal 9 as the information processing apparatus of the present invention.
- the functional block configuration of each device is shown in FIG.
- the server 4 includes an evaluation information acquisition unit 407 instead of the evaluation information acquisition unit 207 with respect to the server 2 in the second embodiment of the present invention, and further includes an action history information acquisition unit 411, The difference is that an action history conversion rule storage unit 412 and an action history information conversion unit 413 are provided.
- the server 4 is configured by a computer apparatus 2600 shown in FIG.
- the action history information acquisition unit 411 includes a network interface 2605 and a CPU 2601 that reads a computer program stored in the ROM 2603 or the storage device 2604 into the RAM 2602 and executes it.
- the action history information conversion unit 413 includes a CPU 2601 that reads a computer program stored in the ROM 2603 or the storage device 2604 into the RAM 2602 and executes the computer program.
- the action history conversion rule storage unit 412 is configured by the storage device 2604.
- the terminal 9 includes an action history information collection unit 903 in addition to the same configuration as that of the terminal 8 in the second embodiment of the present invention.
- the terminal 9 is configured by a computer apparatus 2700 shown in FIG.
- the action history information collection unit 903 includes a network interface 2707 and a CPU 2701 that reads a computer program stored in the ROM 2703 or the storage device 2704 into the RAM 2702 and executes the computer program.
- each functional block constituting each device is not limited to the above-described configuration.
- the behavior history information acquisition unit 411 receives behavior history information representing the history of behavior taken by the browsing user from the terminal 9 for each case information as an inference result presented at the terminal 9.
- the action history information acquisition unit 411 includes, as the action history information, a browsing time of the inference result presentation screen in the terminal 9, an access record for the URL of the case information indicated in the inference result, a browsing time of the screen indicated by the URL, You may acquire the input operation log
- the action history conversion rule storage unit 412 stores action history conversion rules for converting action history information related to each case information into evaluation information that evaluates the degree of knowledge of the user with respect to the case information.
- An example of information stored in the action history conversion rule storage unit 412 is shown in FIG.
- the action history conversion rule indicated by each line includes a condition for the action history information and evaluation information.
- the action history conversion rule on the first line indicates that the action history information is converted into the evaluation information “view” if the browsing time for the detailed page of the case information as the inference result is 10 seconds or more.
- the action history conversion rule on the second line has a browsing time of 100 seconds or more for the detailed page of the case information as the inference result, and there is an input operation to the verification result input text area provided on the page.
- the action history information is converted into evaluation information “verification performed”.
- the action history conversion rule on the third line converts the action history information into evaluation information “pointing input” if there is an input operation to the comment input text area provided on the detailed page of the case information as the inference result. Represents what to do. Also, the action history conversion rule on the fourth line has a viewing time of 10 seconds or more with respect to the detailed page of the case information as the inference result, and if there is an access operation to the attached file with a link attached to the page, This indicates that the action history information is converted into evaluation information “detailed check”.
- FIG. 23 shows an example of information stored in the knowledge level change rule storage unit 208 when the action history conversion rule storage unit 412 stores the information shown in FIG.
- the knowledge level change rules shown in FIG. 23 represent knowledge level change rules corresponding to the evaluation information “browsing”, “verification execution”, “pointing out input”, and “detail check”, respectively.
- the behavior history information conversion unit 413 applies the behavior history conversion rule stored in the behavior history conversion rule storage unit 412 to the behavior history information acquired by the behavior history information acquisition unit 411, thereby evaluating the behavior history information as evaluation information. Convert to
- the evaluation information acquisition unit 407 acquires the evaluation information from the behavior history information conversion unit 413 instead of acquiring the evaluation information selected by the browsing user's input operation.
- the action history information collection unit 903 acquires information representing an operation history for the information input / output unit 802 as action history information. If the information input / output unit 802 is configured by a web browser application, the behavior history information collection unit 903 may acquire a page transition history and an operation history in the web browser as behavior history information. Then, the action history information collection unit 903 transmits the collected action history information to the server 4.
- FIG. 24 is a flowchart for explaining the knowledge level update operation of the information processing system according to the fourth embodiment of the present invention. Since the inference result presentation operation of the information processing system 40 is the same as that of the information processing system 20 according to the second embodiment of the present invention, detailed description in the fourth embodiment is omitted.
- the action history information collection unit 903 of the terminal 9 collects user action history information for each case information as an inference result, and transmits it to the server 4 (step S60).
- the action history information collection unit 903 may acquire the operation history information of the user for the application software presenting the inference result as the action history information.
- the action history information acquisition unit 411 of the server 4 receives action history information from the terminal 9 (step S61).
- the behavior history information conversion unit 413 searches the behavior history conversion rule storage unit 412 for applicable behavior history conversion rules for the behavior history information acquired in step S61 (step S62).
- the action history information conversion unit 413 converts the action history information into evaluation information according to the searched action history conversion rule (step S63).
- the information processing system 40 operates in the same manner as the information processing system 20 according to the second embodiment of the present invention from step S33 to S35, so that each inference used until the case information is obtained. Update the browsing user's knowledge level for the rule.
- the information processing system 40 ends the knowledge level update operation.
- the server 4 includes an inference rule storage unit 201, a knowledge level storage unit 202, a context information acquisition unit 203, an evaluation information acquisition unit 407, a knowledge level change rule storage unit 208, a case information storage unit 210, and an action history information acquisition unit 411. And part or all of the action history conversion rule storage unit 412 may not be provided. Further, these components may be outside the server 4. Furthermore, some or all of inference rules, knowledge levels, context information, evaluation information, knowledge level change rules, case information, action history information, and action history conversion rules may be given to the server 4 from the outside.
- the information processing apparatus can update the knowledge level for each user with respect to each inference rule without burdening the user with work.
- the reason is that the action history information acquisition unit acquires information representing the user's action history with respect to the presented inference result, and the action history information conversion unit uses the action history conversion rule stored in advance. This is because information is converted into evaluation information. Thereby, the information processing apparatus according to the fourth embodiment obtains user evaluation information for accurately updating the knowledge level for each user regarding the inference rule without depending on the user's input work. This is because it becomes possible.
- the action history information acquisition unit has been described as acquiring the page transition information and operation history information in the application software that presents the inference result as the action history information.
- the action history information acquisition unit in the present invention may be any information as long as it is information that represents the user's action with respect to the presented inference result and can be acquired by the computer device.
- the evaluation information acquisition unit is further provided with the evaluation input in the information input / output unit of the terminal, similarly to the evaluation information acquisition unit in the second or third embodiment of the present invention. Information may be received. And the evaluation information acquisition part in 4th Embodiment employ
- the evaluation information acquisition unit in the fourth embodiment may acquire the evaluation information converted by the action history information conversion unit when the evaluation information is not input in the information input / output unit of the terminal. Good.
- the information processing system including the server and the terminal has been described. However, by realizing each function of the server and the terminal in each embodiment in one computer device.
- the information processing apparatus of the present invention can also be configured.
- the context information is character information extracted from the circuit design document data.
- the context information in each embodiment is any information as long as it is information to be inferred, such as character information, image information, measurement data output from the sensor device, operation log data for the device and application software, etc. It may be.
- the inference rule has been described mainly with an example in which the IF-THEN format is used.
- the inference rule in each embodiment may be in other forms used for deriving an inference result from context information.
- the computer program may be stored in a storage device (storage medium) of a computer device, and the computer program may be read and executed by the CPU.
- the present invention is constituted by a code representing such a computer program or the storage medium described above.
- An inference part that obtains inference results by applying inference rules to the context information; Using the information representing the browsing user browsing the inference result, obtain the knowledge level of the viewing user for each inference rule applied until the inference result is obtained, and based on the acquired knowledge level, An inference result index value calculation unit that calculates an index value that collectively represents the depth of knowledge of the browsing user with respect to the inference result; An inference result presentation unit for presenting the inference result based on the index value calculated by the inference result index value calculation unit; For each inference rule applied until the inference result is obtained, the knowledge level of the browsing user with respect to the inference result presented by the inference result presenting unit is expressed as the knowledge level of the viewing user for the inference rule.
- a knowledge level updating unit that updates the information based on the evaluated evaluation information
- An information processing apparatus comprising: (Appendix 2) The inference unit sets the case information related to each inference rule applied until the inference result is obtained as the inference result, The information processing apparatus according to appendix 1, wherein the inference result index value calculation unit calculates the index value for each case information as the inference result. (Appendix 3) The inference result index value calculation unit is further based on the importance of case information as the inference result, in addition to the knowledge level of the browsing user for each inference rule applied until the inference result is obtained, The information processing apparatus according to attachment 2, which calculates an index value.
- the inference result index value calculation unit further includes the browsing user's knowledge level for the case information as the inference result. Based on the index value, The information processing apparatus according to appendix 2 or appendix 3, wherein the knowledge level update unit further updates the knowledge level of the browsing user with respect to the case information based on the evaluation information. (Appendix 5) The inference result index value calculation unit is further based on the occurrence probability of the event inferred by each inference rule, in addition to the knowledge level of the browsing user for each inference rule applied until the inference result is obtained, The information processing apparatus according to any one of supplementary notes 1 to 4, which calculates the index value.
- the inference result index value calculation unit calculates the index value based on an integrated value of the knowledge level of the browsing user with respect to each inference rule applied until the inference result is obtained.
- the information processing apparatus as described in any one.
- (Appendix 7) The information processing apparatus according to any one of Supplementary Note 1 to Supplementary Note 6, wherein the knowledge level update unit executes the update using an increase / decrease value associated with a type of the evaluation information.
- (Appendix 8) An action history for converting the action history information into the evaluation information by applying an action history conversion rule to the action history information representing the action history of the browsing user who has browsed the inference result presented by the inference result presentation unit.
- the information processing apparatus according to any one of supplementary notes 1 to 7, further comprising an information conversion unit.
- the information processing apparatus according to appendix 8, wherein the action history information is an operation history for application software that presents the inference result.
- the knowledge level is a value included in a predetermined range;
- the information processing apparatus according to any one of Supplementary Note 1 to Supplementary Note 9, wherein the knowledge level update unit updates the knowledge level within the predetermined range.
- Appendix 11 The information processing apparatus according to any one of Supplementary Note 1 to Supplementary Note 10, A terminal that collects context information and transmits it to the information processing device, outputs an inference result presented from the information processing device to an output device, and transmits the evaluation information input from the input device to the information processing device; , Information processing system with (Appendix 12) When the information processing apparatus is described in appendix 8 or appendix 9, The information processing system according to appendix 11, wherein the terminal further collects the action history information for the output of the inference result and transmits the collected action history information to the information processing apparatus.
- the inference result is obtained by applying inference rules stored in advance to the input context information, For each inference rule applied until the inference result is obtained, a value stored in advance as a knowledge level representing the depth of knowledge possessed by the browsing user viewing the inference result is acquired, and based on each acquired knowledge level Calculating an index value that comprehensively represents the depth of knowledge of the browsing user with respect to the inference result, Presenting the inference result based on the index value; For the inference result presented, obtain evaluation information that evaluates the degree of knowledge that the browsing user has, An information processing method for updating, based on the evaluation information, a value stored as the knowledge level of the browsing user for the inference rule for each inference rule applied until the inference result is obtained.
- An inference result index value calculation process for calculating an index value that comprehensively represents the depth of knowledge of the browsing user with respect to the inference result, based on each knowledge level;
- An inference result presentation process for presenting the inference result based on the index value;
- an evaluation information acquisition process for acquiring evaluation information in which the degree of knowledge of the browsing user is evaluated;
- a knowledge level update process for updating a value stored in a storage device as a knowledge level of the browsing user for the inference rule based on the evaluation information; Is a computer program that causes a computer device to execute.
- Information processing device 2 3, 4 Server 20, 30, 40 Information processing system 8, 9 Terminal 101, 201, 301 Inference rule storage unit 102, 202, 302 Knowledge level storage unit 103, 203 Context information acquisition unit 104, 204 Inference unit 105, 205, 305 Inference result index value calculation unit 106, 206 Inference result presentation unit 107, 207, 407 Evaluation information acquisition unit 108, 208 Knowledge level change rule storage unit 109, 209, 309 Knowledge level update unit 210, 310 Case information storage unit 411 Action history information acquisition unit 412 Action history conversion rule storage unit 413 Action history information conversion unit 801 Context information collection unit 802 Information input / output unit 903 Action history information collection unit 2500, 2600, 2700 Computer apparatus 2501, 260 , 2701 CPU 2502, 2602, 2702 RAM 2503, 2603, 2703 ROM 2504, 2604, 2704 Storage device 2505, 2705 Input device 2506, 2706 Display device 2605, 2707 Network interface
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
IF 温度>100 & 液体==水
THEN 液体→(変化)→気体・・・(1)
式(1)の1行目は、「温度が摂氏100度以上であり、かつ、液体が水である場合」という条件(IF情報)を表している。また、式(1)の2行目は、IF情報が満たされる場合に「液体が気体に変化する」という事象(THEN情報)を表している。このような推論ルールを用いる情報処理装置は、入力されたコンテキスト情報がIF情報を満たす場合に、該IF情報に対して設定されたTHEN情報が発生すると推論する。
本発明の第1の実施の形態に係る情報処理装置1の機能ブロック構成を図1に示す。図1において、情報処理装置1は、推論ルール記憶部101と、知識レベル記憶部102と、コンテキスト情報取得部103と、推論部104と、推論結果指標値算出部105と、推論結果提示部106と、評価情報取得部107と、知識レベル更新部109とを備える。次に、情報処理装置1のハードウェア構成を図25に示す。図25において、情報処理装置1は、CPU(Central Processing Unit)2501と、RAM(Random Access Memory)2502と、ROM(Read Only Memory)2503と、ハードディスク等の記憶装置2504と、入力装置2505と、表示装置2506とを備えたコンピュータ装置2500によって構成されている。また、推論ルール記憶部101および知識レベル記憶部102は、記憶装置2504によって構成される。また、コンテキスト情報取得部103および評価情報取得部107は、入力装置2505と、ROM2503または記憶装置2504に記憶されたコンピュータ・プログラムをRAM2502に読み込んで実行するCPU2501とによって構成される。また、推論部104と、推論結果指標値算出部105と、知識レベル更新部109とは、ROM2503または記憶装置2504に記憶されたコンピュータ・プログラムをRAM2502に読み込んで実行するCPU2501によって構成される。また、推論結果提示部106は、表示装置2506と、ROM2503または記憶装置2504に記憶されたコンピュータ・プログラムをRAM2502に読み込んで実行するCPU2501とによって構成される。なお、各機能ブロックを構成するハードウェア構成は、上述の構成に限定されない。
次に、本発明の第2の実施の形態について図面を参照して詳細に説明する。第2の実施の形態では、本発明の情報処理装置としてのサーバと、端末とを含む情報処理システムについて説明する。なお、第2の実施の形態の説明において参照する各図面において、本発明の第1の実施の形態と同一の構成および同様に動作するステップには同一の符号を付し、そして、第2の実施の形態における詳細な説明を省略する。
・「閲覧」:推論結果である事例情報に対応付けられたURLに閲覧ユーザがアクセスしたことを表す。つまり、この推論結果としての事例情報に対して、閲覧ユーザは、閲覧により知識を深めたという評価を表す。したがって、この評価情報に対する知識レベル変更ルールは、「閲覧」と評価された事例情報が導出されるまでに適用された各推論ルールに対する閲覧ユーザの知識レベルがある程度深まったとして、それぞれの知識レベルに0.1加算することを表している。
・「検証実施」:そのような推論根拠によってそのような事例情報が導かれることを閲覧ユーザが実際に検証したことを表す。つまり、この推論結果としての事例情報に対して、閲覧ユーザは、検証実施により知識を深めたという評価を表す。したがって、この評価情報に対する知識レベル変更ルールは、「検証実施」と評価された事例情報が導出されるまでに適用された各推論ルールに対する閲覧ユーザの知識がある程度深まったとして、それぞれの知識レベルに0.2加算することを表している。
・「専門領域」:推論結果やその推論根拠について、閲覧ユーザは既に専門的知識を有していることを表す。つまり、この推論結果としての事例情報に対して、閲覧ユーザは、既に詳細な知識を有しているという評価を表す。したがって、この評価情報に対する知識レベル変更ルールは、「専門領域」と評価された事例情報が導出されるまでに適用された各推論ルールに対する閲覧ユーザの知識が元々かなり深かったものとして、それぞれの知識レベルに0.5加算することを表している。
・「見落とし発生」:閲覧ユーザは、そのような推論根拠による推論結果を見落としており、この推論結果を提示されて始めて気がついたことを表す。つまり、この推論結果としての事例情報を見落としてしまう程、閲覧ユーザは、該事例に対して詳しい知識を有していないという評価を表す。したがって、この評価情報に対する知識レベル変更ルールは、「見落とし発生」と評価された事例情報が導出されるまでに適用された各推論ルールに対する閲覧ユーザの知識が浅いとして、それぞれの知識レベルから0.5減算することを表している。
・「不具合埋め込み」:閲覧ユーザがその不具合事例の実施者である、または、その不具合事例と同一の不具合を実際に引き起こしたことを表す。つまり、この推論結果としての不具合を実際に実施してしまう程、閲覧ユーザは、該事例に対して詳しい知識を有していないという評価を表す。したがって、この評価情報に対する知識レベル変更ルールは、「不具合埋め込み」と評価された事例情報が導出されるまでに適用された各推論ルールに対する閲覧ユーザの知識が非常に浅いとして、それぞれの知識レベルから1.0減算することを表している。
次に、本発明の第3の実施の形態について図面を参照して詳細に説明する。なお、第3の実施の形態の説明において参照する各図面において、本発明の第2の実施の形態と同一の構成および同様に動作するステップには同一の符号を付し、そして、第3の実施の形態における詳細な説明を省略する。
次に、本発明の第4の実施の形態について図面を参照して詳細に説明する。なお、第4の実施の形態の説明において参照する各図面において、本発明の第3の実施の形態と同一の構成および同様に動作するステップには同一の符号を付し、そして、第4の実施の形態における詳細な説明を省略する。
(付記1)
コンテキスト情報に対して推論ルールを適用することにより推論結果を得る推論部と、
前記推論結果を閲覧する閲覧ユーザを表す情報を用いて、前記推論結果が得られるまでに適用された各推論ルールに対する前記閲覧ユーザの知識レベルを取得し、取得した各知識レベルに基づいて、前記推論結果に対する前記閲覧ユーザの知識の深さを統合的に表す指標値を算出する推論結果指標値算出部と、
前記推論結果指標値算出部によって算出された指標値に基づいて前記推論結果を提示する推論結果提示部と、
前記推論結果が得られるまでに適用された各推論ルールについて、該推論ルールに対する前記閲覧ユーザの知識レベルを、前記推論結果提示部によって提示された推論結果に対して前記閲覧ユーザが有する知識の程度が評価された評価情報に基づいて更新する知識レベル更新部と、
を備えた情報処理装置。
(付記2)
前記推論部は、前記推論結果を得るまでに適用した各推論ルールに関連する事例情報を前記推論結果とし、
前記推論結果指標値算出部は、前記推論結果としての各事例情報について前記指標値を算出する付記1に記載の情報処理装置。
(付記3)
前記推論結果指標値算出部は、前記推論結果が得られるまでに適用された各推論ルールに対する前記閲覧ユーザの知識レベルに加えて、前記推論結果としての事例情報の重要度にさらに基づいて、前記指標値を算出する付記2に記載の情報処理装置。
(付記4)
前記推論結果指標値算出部は、前記推論結果が得られるまでに適用された各推論ルールに対する前記閲覧ユーザの知識レベルに加えて、前記推論結果としての事例情報に対する前記閲覧ユーザの知識レベルにさらに基づいて前記指標値を算出し、
前記知識レベル更新部は、さらに、前記事例情報に対する前記閲覧ユーザの知識レベルを、前記評価情報に基づいて更新する付記2または付記3に記載の情報処理装置。
(付記5)
前記推論結果指標値算出部は、前記推論結果が得られるまでに適用された各推論ルールに対する前記閲覧ユーザの知識レベルに加えて、各推論ルールによって推論される事象の発生確率にさらに基づいて、前記指標値を算出する付記1から付記4のいずれか1つに記載の情報処理装置。
(付記6)
前記推論結果指標値算出部は、前記推論結果が得られるまでに適用された各推論ルールに対する前記閲覧ユーザの知識レベルの積算値に基づいて、前記指標値を算出する付記1から付記5のいずれか1つに記載の情報処理装置。
(付記7)
前記知識レベル更新部は、前記評価情報の種別に対応付けられた増減値を用いて前記更新を実行する付記1から付記6のいずれか1つに記載の情報処理装置。
(付記8)
前記推論結果提示部によって提示された推論結果を閲覧した前記閲覧ユーザの行動履歴を表す行動履歴情報に、行動履歴変換ルールを適用することにより、該行動履歴情報を前記評価情報に変換する行動履歴情報変換部をさらに備える付記1から付記7のいずれか1つに記載の情報処理装置。
(付記9)
前記行動履歴情報は、前記推論結果を提示するアプリケーションソフトウェアに対する操作履歴である付記8に記載の情報処理装置。
(付記10)
前記知識レベルは所定範囲に含まれる値であり、
前記知識レベル更新部は、前記所定範囲内で前記知識レベルを更新する付記1から付記9のいずれか1つに記載の情報処理装置。
(付記11)
付記1から付記10のいずれか1つに記載の情報処理装置と、
コンテキスト情報を収集して前記情報処理装置に送信し、前記情報処理装置から提示される推論結果を出力装置に出力し、入力装置から入力される前記評価情報を前記情報処理装置に送信する端末と、
を備えた情報処理システム。
(付記12)
前記情報処理装置が付記8または付記9に記載されたものであるとき、
前記端末は、さらに、前記推論結果の出力に対する前記行動履歴情報を収集して前記情報処理装置に送信する付記11に記載の情報処理システム。
(付記13)
入力されたコンテキスト情報に対して、あらかじめ記憶した推論ルールを適用することにより推論結果を求め、
前記推論結果を得るまでに適用した各推論ルールに対して、前記推論結果を閲覧する閲覧ユーザが有する知識の深さを表す知識レベルとしてあらかじめ記憶した値を取得し、取得した各知識レベルに基づいて、前記推論結果に対する前記閲覧ユーザの知識の深さを統合的に表す指標値を算出し、
前記指標値に基づいて前記推論結果を提示し、
提示した推論結果に対して、前記閲覧ユーザが有する知識の程度が評価された評価情報を取得し、
前記推論結果を得るまでに適用した各推論ルールについて、該推論ルールに対する前記閲覧ユーザの知識レベルとして記憶した値を、前記評価情報に基づいて更新する、情報処理方法。
(付記14)
コンテキスト情報を取得するコンテキスト情報取得処理と、
前記コンテキスト情報に対して、あらかじめ記憶装置に記憶された推論ルールを適用することにより推論結果を求める推論処理と、
前記推論結果が得られるまでに適用された各推論ルールに対して、前記推論結果を閲覧する閲覧ユーザが有する知識の深さを表す知識レベルとして記憶装置にあらかじめ記憶された値を取得し、取得した各知識レベルに基づいて、前記推論結果に対する前記閲覧ユーザの知識の深さを統合的に表す指標値を算出する推論結果指標値算出処理と、
前記指標値に基づいて前記推論結果を提示する推論結果提示処理と、
前記推論結果提示処理で提示された推論結果に対して、前記閲覧ユーザが有する知識の程度が評価された評価情報を取得する評価情報取得処理と、
前記推論結果が得られるまでに適用された各推論ルールについて、該推論ルールに対する前記閲覧ユーザの知識レベルとして記憶装置に記憶された値を、前記評価情報に基づいて更新する知識レベル更新処理と、
をコンピュータ装置に実行させるコンピュータ・プログラム。
2、3、4 サーバ
20、30、40 情報処理システム
8、9 端末
101、201、301 推論ルール記憶部
102、202、302 知識レベル記憶部
103、203 コンテキスト情報取得部
104、204 推論部
105、205、305 推論結果指標値算出部
106、206 推論結果提示部
107、207、407 評価情報取得部
108、208 知識レベル変更ルール記憶部
109、209、309 知識レベル更新部
210、310 事例情報記憶部
411 行動履歴情報取得部
412 行動履歴変換ルール記憶部
413 行動履歴情報変換部
801 コンテキスト情報収集部
802 情報入出力部
903 行動履歴情報収集部
2500、2600、2700 コンピュータ装置
2501、2601、2701 CPU
2502、2602、2702 RAM
2503、2603、2703 ROM
2504、2604、2704 記憶装置
2505、2705 入力装置
2506、2706 表示装置
2605、2707 ネットワークインタフェース
Claims (10)
- コンテキスト情報に対して推論ルールを適用することにより推論結果を得る推論部と、
前記推論結果を閲覧する閲覧ユーザを表す情報を用いて、前記推論結果が得られるまでに適用された各推論ルールに対する前記閲覧ユーザの知識レベルを取得し、取得した各知識レベルに基づいて、前記推論結果に対する前記閲覧ユーザの知識の深さを統合的に表す指標値を算出する推論結果指標値算出部と、
前記推論結果指標値算出部によって算出された指標値に基づいて前記推論結果を提示する推論結果提示部と、
前記推論結果が得られるまでに適用された各推論ルールについて、該推論ルールに対する前記閲覧ユーザの知識レベルを、前記推論結果提示部によって提示された推論結果に対して前記閲覧ユーザが有する知識の程度が評価された評価情報に基づいて更新する知識レベル更新部と、
を備えた情報処理装置。 - 前記推論部は、前記推論結果を得るまでに適用した各推論ルールに関連する前記事例情報を前記推論結果とし、
前記推論結果指標値算出部は、前記推論結果としての各事例情報について前記指標値を算出する請求項1に記載の情報処理装置。 - 前記推論結果指標値算出部は、前記推論結果が得られるまでに適用された各推論ルールに対する前記閲覧ユーザの知識レベルに加えて、前記推論結果としての事例情報の重要度にさらに基づいて、前記指標値を算出する請求項2に記載の情報処理装置。
- 前記推論結果指標値算出部は、前記推論結果が得られるまでに適用された各推論ルールに対する前記閲覧ユーザの知識レベルに加えて、前記推論結果としての事例情報に対する前記閲覧ユーザの知識レベルにさらに基づいて前記指標値を算出し、
前記知識レベル更新部は、さらに、前記事例情報に対する前記閲覧ユーザの知識レベルを、前記評価情報に基づいて更新する請求項2または請求項3に記載の情報処理装置。 - 前記推論結果指標値算出部は、前記推論結果が得られるまでに適用された各推論ルールに対する前記閲覧ユーザの知識レベルに加えて、各推論ルールによって推論される事象の発生確率にさらに基づいて、前記指標値を算出する請求項1から請求項4のいずれか1項に記載の情報処理装置。
- 前記知識レベル更新部は、前記評価情報の種別に対応付けられた増減値を用いて前記更新を実行する請求項1から請求項5のいずれか1項に記載の情報処理装置。
- 前記推論結果提示部によって提示された推論結果を閲覧した前記閲覧ユーザの行動履歴を表す行動履歴情報に、行動履歴変換ルールを適用することにより、該行動履歴情報を前記評価情報に変換する行動履歴情報変換部をさらに備える請求項1から請求項6のいずれか1項に記載の情報処理装置。
- 請求項1から請求項7のいずれか1項に記載の情報処理装置と、
コンテキスト情報を収集して前記情報処理装置に送信し、前記情報処理装置から提示される推論結果を出力装置に出力し、入力装置から入力される前記評価情報を前記情報処理装置に送信する端末と、
を備えた情報処理システム。 - 入力されたコンテキスト情報に対して、あらかじめ記憶した推論ルールを適用することにより推論結果を求め、
前記推論結果を得るまでに適用した各推論ルールに対して、前記推論結果を閲覧する閲覧ユーザが有する知識の深さを表す知識レベルとしてあらかじめ記憶した値を取得し、取得した各知識レベルに基づいて、前記推論結果に対する前記閲覧ユーザの知識の深さを統合的に表す指標値を算出し、
前記指標値に基づいて前記推論結果を提示し、
提示した推論結果に対して、前記閲覧ユーザが有する知識の程度が評価された評価情報を取得し、
前記推論結果を得るまでに適用した各推論ルールについて、該推論ルールに対する前記閲覧ユーザの知識レベルとして記憶した値を、前記評価情報に基づいて更新する、情報処理方法。 - コンテキスト情報を取得するコンテキスト情報取得処理と、
前記コンテキスト情報に対して、あらかじめ記憶装置に記憶された推論ルールを適用することにより推論結果を求める推論処理と、
前記推論結果が得られるまでに適用された各推論ルールに対して、前記推論結果を閲覧する閲覧ユーザが有する知識の深さを表す知識レベルとして記憶装置にあらかじめ記憶された値を取得し、取得した各知識レベルに基づいて、前記推論結果に対する前記閲覧ユーザの知識の深さを統合的に表す指標値を算出する推論結果指標値算出処理と、
前記指標値に基づいて前記推論結果を提示する推論結果提示処理と、
前記推論結果提示処理で提示された推論結果に対して、前記閲覧ユーザが有する知識の程度が評価された評価情報を取得する評価情報取得処理と、
前記推論結果が得られるまでに適用された各推論ルールについて、該推論ルールに対する前記閲覧ユーザの知識レベルとして記憶装置に記憶された値を、前記評価情報に基づいて更新する知識レベル更新処理と、
をコンピュータ装置に実行させるコンピュータ・プログラム。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2013549118A JP6079638B2 (ja) | 2011-12-15 | 2012-12-12 | 情報処理装置、情報処理システム、情報処理方法、および、コンピュータ・プログラム |
US14/364,373 US9886666B2 (en) | 2011-12-15 | 2012-12-12 | Information processing device, information processing system, information processing method and computer-readable medium |
EP12857998.4A EP2793144A4 (en) | 2011-12-15 | 2012-12-12 | INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING METHOD, AND COMPUTER PROGRAM |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011274792 | 2011-12-15 | ||
JP2011-274792 | 2011-12-15 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2013088708A1 true WO2013088708A1 (ja) | 2013-06-20 |
Family
ID=48612182
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2012/007930 WO2013088708A1 (ja) | 2011-12-15 | 2012-12-12 | 情報処理装置、情報処理システム、情報処理方法、および、コンピュータ・プログラム |
Country Status (4)
Country | Link |
---|---|
US (1) | US9886666B2 (ja) |
EP (1) | EP2793144A4 (ja) |
JP (1) | JP6079638B2 (ja) |
WO (1) | WO2013088708A1 (ja) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013088708A1 (ja) * | 2011-12-15 | 2013-06-20 | 日本電気株式会社 | 情報処理装置、情報処理システム、情報処理方法、および、コンピュータ・プログラム |
CN111064433A (zh) | 2018-10-17 | 2020-04-24 | 太阳能安吉科技有限公司 | 光伏系统故障和警报 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS62281051A (ja) * | 1986-05-30 | 1987-12-05 | Nissan Motor Co Ltd | 車両用故障診断装置 |
JPS6366639A (ja) * | 1986-09-08 | 1988-03-25 | Nec Corp | 知識ベ−ス・システムにおける柔軟説明方式 |
JPH02201295A (ja) * | 1989-01-31 | 1990-08-09 | Toshiba Corp | プラント機器故障診断作業支援装置 |
JPH0535479A (ja) * | 1991-07-29 | 1993-02-12 | Ricoh Co Ltd | 情報処理装置 |
JP2007280301A (ja) | 2006-04-11 | 2007-10-25 | Omron Corp | 不具合管理装置、不具合管理方法、不具合管理プログラム、およびこれを記録した記録媒体 |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2845508B2 (ja) | 1989-08-31 | 1999-01-13 | 株式会社東芝 | エキスパートシステムの推論機構 |
US7853485B2 (en) | 2005-11-22 | 2010-12-14 | Nec Laboratories America, Inc. | Methods and systems for utilizing content, dynamic patterns, and/or relational information for data analysis |
KR100976722B1 (ko) * | 2007-12-18 | 2010-08-18 | 한국과학기술정보연구원 | 사용자 맞춤형 연구 정보 제공 방법 및 시스템 |
JP5637143B2 (ja) * | 2009-11-05 | 2014-12-10 | 日本電気株式会社 | 共同開発支援システム、共同開発支援方法及びプログラム |
JPWO2012060105A1 (ja) * | 2010-11-05 | 2014-05-12 | 日本電気株式会社 | 情報処理装置 |
WO2013088708A1 (ja) * | 2011-12-15 | 2013-06-20 | 日本電気株式会社 | 情報処理装置、情報処理システム、情報処理方法、および、コンピュータ・プログラム |
-
2012
- 2012-12-12 WO PCT/JP2012/007930 patent/WO2013088708A1/ja active Application Filing
- 2012-12-12 JP JP2013549118A patent/JP6079638B2/ja active Active
- 2012-12-12 US US14/364,373 patent/US9886666B2/en active Active
- 2012-12-12 EP EP12857998.4A patent/EP2793144A4/en not_active Withdrawn
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS62281051A (ja) * | 1986-05-30 | 1987-12-05 | Nissan Motor Co Ltd | 車両用故障診断装置 |
JPS6366639A (ja) * | 1986-09-08 | 1988-03-25 | Nec Corp | 知識ベ−ス・システムにおける柔軟説明方式 |
JPH02201295A (ja) * | 1989-01-31 | 1990-08-09 | Toshiba Corp | プラント機器故障診断作業支援装置 |
JPH0535479A (ja) * | 1991-07-29 | 1993-02-12 | Ricoh Co Ltd | 情報処理装置 |
JP2007280301A (ja) | 2006-04-11 | 2007-10-25 | Omron Corp | 不具合管理装置、不具合管理方法、不具合管理プログラム、およびこれを記録した記録媒体 |
Non-Patent Citations (2)
Title |
---|
NAOKI OHSUGI ET AL.: "Software Function Recommender System Based on Collaborative Filtering", TRANSACTION OF INFORMATION PROCESSING SOCIETY OF JAPAN, vol. 45, no. 1, 2004 |
See also references of EP2793144A4 |
Also Published As
Publication number | Publication date |
---|---|
EP2793144A1 (en) | 2014-10-22 |
US20140358827A1 (en) | 2014-12-04 |
JPWO2013088708A1 (ja) | 2015-04-27 |
US9886666B2 (en) | 2018-02-06 |
EP2793144A4 (en) | 2016-12-21 |
JP6079638B2 (ja) | 2017-02-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wang et al. | Comparison of ARIMA and GM (1, 1) models for prediction of hepatitis B in China | |
US11238233B2 (en) | Artificial intelligence engine for generating semantic directions for websites for automated entity targeting to mapped identities | |
US8893287B2 (en) | Monitoring and managing user privacy levels | |
KR20110069734A (ko) | 리스크 정보의 분산 추출 및 집성을 위한 시스템 및 방법 | |
Bornmann et al. | The proposal of a broadening of perspective in evaluative bibliometrics by complementing the times cited with a cited reference analysis | |
Dasu | Data glitches: Monsters in your data | |
CN106796618A (zh) | 时序预测装置和时序预测方法 | |
WO2020004154A1 (ja) | 情報処理装置、情報処理方法及びプログラム | |
WO2011065295A1 (ja) | 評判分析装置、評判分析方法、および評判分析用プログラム | |
Gitzel | Data Quality in Time Series Data: An Experience Report. | |
JP2021043939A (ja) | 自動難度レベル推定のシステム及び方法 | |
JP6079638B2 (ja) | 情報処理装置、情報処理システム、情報処理方法、および、コンピュータ・プログラム | |
JP2012094056A (ja) | ユーザ状態推定システム、ユーザ状態推定方法及びユーザ状態推定プログラム | |
JP4447552B2 (ja) | 情報提供方法及び装置及びプログラム及びコンピュータ読み取り可能な記録媒体 | |
JP2012510662A (ja) | ウェブページの接続時間及び訪問度に基づいたウェブ検索システム及びその方法 | |
JP2015036923A (ja) | 評価集計装置、評価順位作成装置、評価集計方法及びプログラム | |
Wlodarczyk et al. | Current trends in predictive analytics of big data | |
Dalvi et al. | Entropy analysis for identifying significant parameters for seismic soil liquefaction | |
CN109582802B (zh) | 一种实体嵌入方法、装置、介质及设备 | |
Khan et al. | Privacy preserved and decentralized smartphone recommendation system | |
Colbaugh et al. | Emerging topic detection for business intelligence via predictive analysis of'meme'dynamics | |
JP2022111544A (ja) | 情報処理システム、及び情報処理方法 | |
WO2024004384A1 (ja) | 情報処理装置及び情報処理方法、並びにコンピュータプログラム | |
US20230112763A1 (en) | Generating and presenting a text-based graph object | |
JP6833235B1 (ja) | 情報処理システム、及び情報処理方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 12857998 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2013549118 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14364373 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2012857998 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |