WO2016006021A1 - Dispositif d'analyse de données, procédé de commande pour dispositif d'analyse de données, et programme de commande pour dispositif d'analyse de données - Google Patents

Dispositif d'analyse de données, procédé de commande pour dispositif d'analyse de données, et programme de commande pour dispositif d'analyse de données Download PDF

Info

Publication number
WO2016006021A1
WO2016006021A1 PCT/JP2014/068004 JP2014068004W WO2016006021A1 WO 2016006021 A1 WO2016006021 A1 WO 2016006021A1 JP 2014068004 W JP2014068004 W JP 2014068004W WO 2016006021 A1 WO2016006021 A1 WO 2016006021A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
relationship
driver
predetermined case
unit
Prior art date
Application number
PCT/JP2014/068004
Other languages
English (en)
Japanese (ja)
Inventor
守本　正宏
池上　成朝
秀樹 武田
彰晃 花谷
Original Assignee
株式会社Ｕｂｉｃ
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社Ｕｂｉｃ filed Critical 株式会社Ｕｂｉｃ
Priority to PCT/JP2014/068004 priority Critical patent/WO2016006021A1/fr
Publication of WO2016006021A1 publication Critical patent/WO2016006021A1/fr

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
    • B60R16/00—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for
    • B60R16/02—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
    • B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems

Definitions

  • the present invention relates to a data analysis apparatus and the like that can extract data related to a predetermined case from a plurality of data acquired from around the vehicle.
  • Patent Document 1 discloses a driving support system that allows a driver to visually obtain driving support information regardless of the direction of the head or line of sight.
  • Patent Document 2 discloses a lane departure warning system that can prompt a driver to keep the vehicle in the center of the lane.
  • Patent Documents 1 and 2 present information that is considered necessary for the driver to drive safely, and estimate the driver's intention based on the result of learning the driver's personality. It was not possible.
  • the present invention has been made in view of the above-described problems, and an object thereof is necessary for a driver by extracting data related to a predetermined case from a plurality of data acquired from around the vehicle. It is to provide a data analysis device or the like that can notify the driver of data.
  • a data analysis apparatus is a data analysis apparatus capable of extracting data related to a predetermined case from a plurality of data acquired from around the vehicle. If undecided data that has not been determined whether or not related to the case is newly acquired, based on the already determined data that has been determined whether or not it is related to the predetermined case by the driver driving the vehicle A relationship evaluation unit that evaluates the relationship between the undetermined data and the predetermined case, and a data notification unit that notifies the driver of the undetermined data according to the relationship evaluated by the relationship evaluation unit. I have.
  • the data analysis apparatus further includes a score calculation unit that calculates a score indicating the strength of the relationship between the predetermined data and the predetermined case, and the relationship evaluation unit includes the undetermined data.
  • the data notification unit When the relationship evaluation unit evaluates that the undetermined data and the predetermined case are related, the undetermined data can be notified to the driver.
  • the data analysis apparatus further includes an element evaluation unit that evaluates each data element included in the already determined data based on a predetermined criterion, and the score calculation unit is evaluated by the element evaluation unit. The score can be calculated using the obtained results.
  • the data analysis apparatus uses the result evaluated by the element evaluation unit, and the score calculated by the score calculation unit as an index indicating the relationship between the already determined data and the predetermined case Among these, a threshold specifying unit that specifies a score that can exceed the target value set for the precision as a predetermined threshold may be further included.
  • the data analysis apparatus includes a moving average of scores respectively calculated for a plurality of already-determined data acquired along a time series, and a plurality of pieces acquired along a time series.
  • a condition determination unit that determines the level of correlation with the moving average of the scores calculated for each of the undecided data is further provided, and the relationship evaluation unit determines whether the undecided data is based on the result determined by the condition determination unit. And the relationship between a given case and the case can be evaluated.
  • the data analysis apparatus obtains a result of determination by a driver as to whether or not predetermined data is related to a predetermined case from the driver through a predetermined input unit.
  • a determination data acquisition unit that acquires the determination data can be further provided.
  • the data analysis apparatus includes a relationship adding unit that adds relationship information indicating that undecided data is related to a predetermined case based on a result evaluated by the relationship evaluation unit. Furthermore, it can be provided.
  • the data analysis device includes an external image captured from a vehicle, an external sound recorded from the vehicle, an internal image captured of the driver, an internal sound recorded in a room of the vehicle,
  • the data acquisition part which acquires at least one of the map information of the road where a vehicle is drive
  • the predetermined case may be to avoid a collision of the vehicle with an obstacle.
  • a control method for a data analysis apparatus is a control method for a data analysis apparatus that can extract data related to a predetermined case from a plurality of data acquired from around a vehicle.
  • undetermined data that is not determined whether or not related to a predetermined case is newly acquired, it is determined whether or not it is related to the predetermined case by a driver driving the vehicle
  • the relationship evaluation step for evaluating the relationship between the undetermined data and the predetermined case, and the undetermined data is notified to the driver according to the relationship evaluated in the relationship evaluation step.
  • a data notification step is a control method for a data analysis apparatus that can extract data related to a predetermined case from a plurality of data acquired from around a vehicle.
  • a control program for a data analysis apparatus is a control program for a data analysis apparatus that can extract data related to a predetermined case from a plurality of data acquired from around the vehicle. Whether or not the data analysis device is related to the predetermined case by the driver who drives the vehicle when new data is acquired in the data analysis device that has not been determined whether or not it is related to the predetermined case. Based on the already-determined data that has been determined, the relationship evaluation function that evaluates the relationship between the undetermined data and the predetermined case, and the relationship that has been evaluated by the relationship evaluation function A data notification function for notifying the driver of data is realized.
  • undetermined data that has not been determined whether or not it relates to a predetermined case has been newly acquired.
  • the relationship between the undetermined data and the predetermined case is evaluated based on the already determined data for which it is determined whether or not the driver is related to the predetermined case.
  • FIG. 2 is a schematic diagram showing an outline of the data analysis apparatus 100.
  • the data analysis apparatus 100 is an apparatus that can extract data related to a predetermined case from a plurality of data acquired from around the vehicle.
  • the data analysis device 100 only needs to be a device that can execute the process described below, and can be realized using, for example, an electronic control unit (ECU), a personal computer, a smartphone, or other electronic devices.
  • ECU electronice control unit
  • the data analysis apparatus 100 acquires, for example, an external image (data 1b) taken from a vehicle as undetermined data that has not been determined whether or not it relates to a predetermined case.
  • the “predetermined case” broadly includes cases, objects, situations, actions, and the like that the driver of the vehicle is considered to exhibit information processing ability.
  • obstacles eg, pedestrians, guardrails, other Collision avoidance with a vehicle, etc., garage entry, lane change, joining / leaving to a highway, and the like.
  • the data analysis apparatus 100 determines whether or not a driver (for example, a skilled driver) relates to the predetermined case when undetermined data that has not been determined whether or not related to the predetermined case is newly acquired. Based on the determined already determined data, the relationship between the undetermined data and the predetermined case is evaluated. Specifically, the data analysis apparatus 100 extracts the data element 2 from the data 1b (for example, an external image), and obtains the score 5e of the data 1b from the data element 2 evaluated using the already determined data. calculate. When the calculated score 5e satisfies a predetermined condition (for example, the score 5e exceeds a predetermined threshold), the data analysis apparatus 100 uses the data 1b as a driver (for example, an unskilled driver). Inform.
  • a driver for example, a skilled driver
  • the data analysis apparatus 100 can determine whether or not to notify the driver of new undecided data based on the result of determination by the driver as to whether or not it is related to a predetermined case. For example, when the skilled driver experiences a near-miss, the data analysis apparatus 100 learns the relationship between the near-miss situation (predetermined case) and an external image indicating the situation, and the unskilled driver encounters the same situation. As a result, when a similar external image is acquired, a part of or all of the similar external image or data (for example, notification by voice) processed from the similar external image is notified to the unskilled driver. it can.
  • the data analysis apparatus 100 has an effect that the driver can be notified of data required by the driver (for example, an external image showing a high-risk situation leading to an accident).
  • FIG. 1 is a block diagram showing a main configuration of the data analysis apparatus 100.
  • the data analysis apparatus 100 includes a control unit 10 (a data acquisition unit 11, an already determined data acquisition unit 12, an element evaluation unit 13, a score calculation unit 14, a condition determination unit 15, and a relationship evaluation unit 16.
  • the control unit 10 comprehensively controls various functions of the data analysis apparatus 100.
  • the control unit 10 includes a data acquisition unit 11, an already-determined data acquisition unit 12, an element evaluation unit 13, a score calculation unit 14, a condition determination unit 15, a relationship evaluation unit 16, a relationship assignment unit 17, a data notification unit 18, and a threshold value specification.
  • the unit 19 and the storage unit 20 are included.
  • the data acquisition unit 11 acquires data 1 from around the vehicle.
  • the data acquisition unit 11 is, for example, an external image captured from the vehicle (for example, an image capturing a road extending in the traveling direction of the vehicle), an external sound recorded from the vehicle, and an internal image capturing a driver (for example, Images of the interior of the vehicle including the driver's head), internal audio recorded in the vehicle's interior (for example, audio logs of a master driver, driving recorder, etc.), map information (GPS) on which the vehicle is currently traveling Information), biometric information of the driver (for example, heart rate, blood pressure, blood state, etc.), and operation information given to the vehicle (for example, information for controlling steering, etc.), It can be acquired as data 1.
  • the data acquisition unit 11 can further acquire text data such as accident records, insurance policies, and e-mails.
  • the data analysis apparatus 100 acquires not only structured data (for example, images, biological information, etc.) but also unstructured data (for example, speech, text, etc.) and basic information for comprehensive analysis. can do. Therefore, the data analyzing apparatus 100 can accurately notify the driver of data required by the driver.
  • structured data for example, images, biological information, etc.
  • unstructured data for example, speech, text, etc.
  • the data acquisition unit 11 outputs, to the already-determined data acquisition unit 12 and the element evaluation unit 13, the data 1 a to be determined by the driver as to whether or not the acquired data 1 is related to a predetermined case.
  • Data 1b (undecided data) is output to the score calculation unit 14.
  • the already-determined data acquisition unit 12 acquires the result (review result 5a) determined by the driver as to whether or not the data 1a is related to a predetermined case from the driver via the input unit 40. Data (a pair of data 1a and review result 5a) is acquired. Specifically, the already-determined data acquisition unit 12 acquires the review result 5a corresponding to the data 1a input from the data acquisition unit 11 based on the input information 5b acquired from the input unit 40. Then, the already-determined data acquisition unit 12 outputs the review result 5 a to the element evaluation unit 13 and the threshold specifying unit 19.
  • the driver that gives the review result 5a to the data analysis apparatus 100 and the driver that receives the review result from the data analysis apparatus 100 are the same driver.
  • different drivers may be used.
  • the data analysis apparatus 100 learns the experience / judgment criteria of the skilled driver, and can notify the data 1b to the unskilled driver based on the learning result. That is, the data analysis apparatus 100 can make use of the experience of the skilled driver to the unskilled driver.
  • the element evaluation unit 13 evaluates each data element included in the already determined data based on a predetermined standard. Specifically, when the data 1a is a voice, the element evaluation unit 13 recognizes the voice and converts the voice into characters (document data). Then, the element evaluation unit 13 transmits transmission information representing a dependency relationship between a keyword (data element) included in the document data and a result (review result 5a) determined by the driver with respect to the data 1a (voice) including the keyword. The keyword can be evaluated by calculating the weight of the keyword using the amount as one of the predetermined criteria.
  • the element evaluation unit 13 may recognize the speech using an arbitrary speech recognition algorithm (for example, a hidden Markov model, a Kalman filter, a neural network, or the like).
  • the element evaluation unit 13 uses an arbitrary image recognition technique (for example, a technique such as pattern matching, Bayesian estimation, Markov chain Monte Carlo, or the like) to include an object ( For example, pedestrians, other vehicles, and other obstacles) can be specified as data elements.
  • the element evaluation unit 13 transmits the amount of transmitted information that represents the dependency relationship between the object (data element) included in the image and the result (review result 5a) determined by the driver for the data 1a (image) including the object. Can be evaluated by calculating the weight of the object as one of the predetermined criteria.
  • the element evaluation unit 13 outputs element information 5 c that is a pair of the data element and the weight of the data element to the score calculation unit 14 and the storage unit 20.
  • the score calculation unit 14 uses the result (element information 5c) evaluated by the element evaluation unit 13 to calculate a score 5d indicating the strength of the relationship between the data 1a and the predetermined case.
  • the score calculation unit 14 outputs the calculated score 5d to the threshold specifying unit 19.
  • data 1b undecided data
  • the score calculation unit 14 calculates a score 5e for the data 1b, and outputs the calculated score 5e to the condition determination unit 15.
  • the score calculation unit 14 can calculate the score (score 5d or score 5e) of the data 1 by adding the weights of the data elements included in the data 1 (data 1a or data 1b). For example, let us consider a case where data 1 is an internal voice recorded in a vehicle room, such as “turn right at the next intersection”. In this case, as a result of the evaluation of the data elements “intersection” and “right turn” by the element evaluation unit 13, when the weights “1.2” and “2.2” are set, the score calculation unit 14 The score of the data 1 can be calculated as “3.4” (1.2 + 2.2).
  • the score calculation unit 14 generates an element vector indicating whether or not a predetermined data element is included in the data 1.
  • each element of the element vector takes a value of “0” or “1”, thereby determining whether or not a predetermined data element associated with the element is included in the data 1. It is a vector to show.
  • the score calculation unit 14 changes the element corresponding to the “intersection” of the element vector from “0” to “1”. Then, the score calculation unit 14 calculates the inner product of the element vector (vertical vector) and the weight vector (vertical vector having the weight for each data element as an element) as in the following equation, thereby obtaining the data 1 Score S is calculated.
  • s represents an element vector
  • W represents a weight vector
  • T represents transposing a matrix / vector (replaces rows and columns).
  • the score calculation unit 14 may calculate the score S according to the following formula.
  • m j represents the appearance frequency of the j-th data element
  • w i represents the weight of the i-th data element
  • the score calculation unit 14 calculates the result (weight of the first data element) of the first data element included in the data 1a and / or the data 1b and the second data included in the data 1a and / or the data 1b.
  • the score 5d and / or the score 5e may be calculated based on the result of evaluating the data element (weight of the second data element). That is, when the first data element appears in the data, the score calculation unit 14 also refers to the frequency at which the second data element appears in the data (that is, the correlation or co-occurrence between the first data element and the second data element). ) Can be taken into account.
  • the data analysis apparatus 100 can calculate the score in consideration of the correlation between the data elements, and therefore can extract data related to a predetermined case with higher accuracy.
  • the condition determination unit 15 determines whether the data 1b satisfies a predetermined condition for notifying the driver of the data 1b based on the score 5e calculated by the score calculation unit 14. For example, the condition determination unit 15 compares one of the predetermined conditions to determine whether the score 5e exceeds the conformance threshold 6 by comparing the score 5e with the conformance threshold (predetermined threshold) 6. You may judge as.
  • condition determination part 15 is respectively with respect to the moving average of the score 5d each calculated with respect to the some data 1a acquired along a time series, and the some data 1b acquired along a time series, respectively. Whether the correlation with the moving average of the calculated score 5e has increased may be determined as one of the predetermined conditions. For example, in the case where the review result 5a indicating that the plurality of data 1a are in a situation of experiencing a near-miss is data obtained from an expert driver, the condition determination unit 15 performs each for the plurality of data 1a. The moving average of the calculated score 5d is extracted as a predetermined pattern.
  • the condition determination unit 15 calculates the correlation between the predetermined pattern and the moving average of the score 5e. In other words, the condition determination unit 15 calculates the degree of coincidence (correlation) between the two while shifting the elapsed time and / or score. When the correlation is high, the condition determination unit 15 determines that the current score 5e assumes a similar value so that it will be linked to the predetermined pattern in the future (that is, a similar near-miss is likely to occur). judge.
  • the condition determination unit 15 has a correlation between the change of the biological information (data 1a) of the other driver acquired in the past by the data acquisition unit 11 and the change of the biological information (data 1b) of the driver currently driving. Whether or not it has increased may be determined as one of the predetermined conditions. For example, when the review result 5a indicating that the biometric information (data 1a) is in a situation of experiencing a near-miss is data obtained from a skilled driver, the condition determination unit 15 correlates the transition of both biometric information. When the correlation becomes high, it is determined that the current biological information takes the same value in the future so as to be linked to the past biological information (that is, there is a high possibility that a similar near-miss occurs). . The condition determination unit 15 outputs the determined result (determination result 5f) to the relationship evaluation unit 16.
  • the relationship evaluation unit 16 is related to the predetermined case by the driver who drives the vehicle when new undetermined data (data 1b) for which it is not determined whether or not it is related to the predetermined case is acquired. Based on the already determined data (a pair of the data 1a and the review result 5a) for which it is determined whether or not, the relationship between the undecided data and the predetermined case is evaluated. For example, when the score 5e calculated by the score calculation unit 14 exceeds the threshold 6 as an index indicating the relationship between the undetermined data (data 1b) and a predetermined case (that is, exceeded by the condition determination unit 15) If it is determined that the undetermined data is related to the predetermined case, it is evaluated. The relationship evaluation unit 16 outputs the evaluated result (evaluation result 5 g) to the relationship providing unit 17.
  • the relationship providing unit 17 assigns relationship information 5h indicating that the undetermined data (data 1b) is related to a predetermined case,
  • the relationship information 5 h is output to the data notification unit 18.
  • the data notification unit 18 notifies the driver of undetermined data (data 1b) according to the relationship evaluated by the relationship evaluation unit 16. Specifically, the data notification unit 18 notifies the driver of the data 1b to which the relationship information 5h indicating the relationship with the predetermined case is added by the relationship adding unit 17. The data notification unit 18 may notify the driver of a part of the data 1b or data obtained by processing the data 1b.
  • the data notification unit 18 displays the external image on a monitor provided in the passenger compartment, thereby notifying the driver that the inter-vehicle distance from the preceding vehicle is short. can do.
  • reporting part 18 can alert
  • the threshold value specifying unit 19 can exceed the target value (target adaptation rate) set for the accuracy rate indicating the ratio of the data 1a determined to be related to the predetermined case to the data group including the predetermined number of data.
  • the smallest minimum score is identified as the fitness threshold 6.
  • the threshold specifying unit 19 rearranges the scores 5d in descending order.
  • the threshold value specifying unit 19 scans the review result 5a assigned to the data 1a in order from the data 1a having the maximum score 5d (score rank is first), and “relevant to a predetermined case”.
  • the ratio of the number of pieces of data to which the review result 5a is given to the number of pieces of data that have been scanned at the present time (matching rate) is sequentially calculated.
  • the threshold value specifying unit 19 calculates the matching rate as 0.9 (18/20).
  • the threshold specifying unit 19 calculates the precision as 0.875 (35/40).
  • the threshold value specifying unit 19 calculates all the precisions for the data 1a and specifies the minimum score that can exceed the target precision. Specifically, the threshold specifying unit 19 scans the precision calculated for the data 1a in order from the data 1a having the minimum score 5d (score rank is 100th), and the precision is the target. When the precision is exceeded, the score corresponding to the precision is output to the condition determination unit 15 and the storage unit 20 as the minimum score (fit threshold 6) that can maintain the target precision.
  • the storage unit 20 associates the data element included in the element information 5c with the result (weight) of the evaluation of the data element, and stores the storage unit 30. To store. Thereby, the data analysis apparatus 100 can extract data related to a predetermined case by analyzing the current data based on the result of analyzing the past data (weight as a result of evaluating the data element). .
  • the storage unit 20 stores the adaptation threshold 6 in the storage unit 30.
  • the input unit (predetermined input unit) 40 receives input from the driver.
  • FIG. 1 shows a configuration in which the data analysis device 100 includes an input unit 40 (for example, a configuration in which a keyboard, a mouse, and the like are connected as the input unit 40).
  • the input unit 40 communicates with the data analysis device 100. It may be an external input device (for example, a client terminal) that is connected as possible.
  • the storage unit (predetermined storage unit) 30 is a storage device configured by an arbitrary recording medium such as a hard disk, an SSD (silicon state drive), a semiconductor memory, a DVD, and the like.
  • a control program capable of controlling the data analysis apparatus 100 is stored. 1 illustrates a configuration in which the data analysis device 100 includes the storage unit 30, the storage unit 30 may be an external storage device connected to the data analysis device 100 so as to be communicable.
  • the already-determined data acquisition unit 12 can receive feedback on the determination from the driver. That is, the driver can input as feedback whether the result determined by the data analysis apparatus 100 is valid or not.
  • the element evaluation unit 13 can re-evaluate each data element based on the feedback. Specifically, the element evaluation unit 13 calculates the weight of each data element according to the following formula.
  • w i, L represents the weight of the i-th data element after the L-th learning
  • ⁇ L represents a learning parameter in the L-th learning
  • ⁇ represents a learning effect threshold
  • the element evaluation unit 13 can recalculate the weight based on the newly obtained feedback with respect to the determination of the data analysis apparatus 100.
  • the data analysis apparatus 100 can obtain a weight suitable for the data to be analyzed, and can accurately calculate a score based on the weight, so that data related to a predetermined case can be extracted with higher accuracy. .
  • FIG. 3 is a schematic diagram showing the effect of the data analysis apparatus 100.
  • FIG. 3A shows that the response to a sudden jump of a pedestrian differs between an expert driver and an unskilled driver
  • FIG. This shows that rear-end collisions due to carelessness and misjudgment can be avoided by experience of skilled drivers.
  • the data analysis apparatus 100 learns the relationship between the near-miss situation (predetermined case) and data indicating the situation, and the unskilled driver is the same.
  • the similar data can be notified to an unskilled driver.
  • the data analysis apparatus 100 can cause the non-experienced driver to perform the same action as the skilled driver in the situation as shown in (a) and (b) of FIG.
  • FIG. 4 is a schematic diagram showing a further effect produced by the data analysis apparatus 100.
  • FIG. 4A shows that maintenance costs (periodic maintenance, car insurance, etc.) are generated even if the vehicle is not used (data).
  • the processing executed by the data analysis device 100 (the control method of the data analysis device 100) is performed when the undetermined data (data 1b) that has not been determined whether or not it relates to a predetermined case is newly acquired.
  • the relationship between the undecided data and the predetermined case is determined. It includes a relationship evaluation step to be evaluated, and a data notification step of notifying the driver of undecided data according to the relationship evaluated in the relationship evaluation step.
  • FIG. 5 is a detailed flowchart showing an example of processing executed by the data analysis apparatus 100.
  • parenthesized “ ⁇ steps” represent steps included in the control method of the data analysis apparatus.
  • the data acquisition unit 11 acquires data 1a to be determined by the driver as to whether or not it is related to a predetermined case (for example, from a camera that captures an external image, a microphone that records internal sound, etc.) (step 1).
  • a predetermined case for example, from a camera that captures an external image, a microphone that records internal sound, etc.
  • step is abbreviated as “S”.
  • the already-determined data acquisition unit 12 acquires the result (review result 5a) determined by the driver as to whether or not the data 1a is related to a predetermined case via the input unit 40 (S2).
  • the element evaluation unit 13 evaluates each data element included in the data determined by the driver as to whether or not it is related to the predetermined case based on a predetermined criterion (S3).
  • the score calculation unit 14 calculates, for each data 1a, a score 5d indicating the strength of the relationship with the predetermined case based on the result (element information 5c) evaluated by the element evaluation unit 13 (S4). ),
  • the threshold value specifying unit 19 sets a target value (target compliance rate) set for the accuracy rate indicating the ratio of the data 1a determined to be related to the predetermined case to the data group including the predetermined number of data.
  • the minimum score that can exceed is specified as the matching threshold 6 (S5).
  • the score calculation unit 14 calculates a score 5e indicating the strength of the relationship with the predetermined case for each data 1b based on the result (element information 5c) evaluated by the element evaluation unit 13 ( S6). Based on the result (element information 5c) evaluated by the element evaluation unit 13, the condition determination unit 15 has a score 5e calculated for the data 1b that has not yet been determined whether or not it is related to the predetermined case. It is determined whether or not the conformity threshold 6 has been exceeded (S7), and if it is determined that it has been exceeded (YES in S7), the relationship evaluation unit 16 determines that the data 1b is related to the predetermined case. (S8, relationship evaluation step).
  • the relationship assigning unit 17 assigns relationship information (review result by the document analysis system 100) indicating that the data 1b is related to the predetermined case to the data 1b evaluated by the relationship evaluating unit 16 (S9). ). Finally, the data notification unit 18 notifies the driver of the data 1b (S10, data notification step).
  • control method may optionally include not only the above-described processing described with reference to FIG. 5 but also processing executed in each unit included in the control unit 10.
  • the data analyzing apparatus 100 determines whether or not the driver is related to the predetermined case when the undetermined data that has not been determined whether or not related to the predetermined case is newly acquired. Based on the determined data, the relationship between the undecided data and the predetermined case is evaluated, and the determined data is notified to the driver according to the relationship.
  • the data analysis apparatus 100 has an effect that it can notify the driver of data required by the driver.
  • the data analysis device of the present invention can function as a server device that is communicably connected to a user terminal via a network.
  • the server device has the same effect as the data analyzing device 100 when the data analyzing device 100 provides a function.
  • the control block (particularly, the control unit 10) of the data analysis apparatus 100 may be realized by a logic circuit (hardware) formed in an integrated circuit (IC chip) or the like, or using a CPU (Central Processing Unit). It may be realized by software.
  • the data analysis apparatus 100 has a CPU that executes instructions of a control program of the data analysis apparatus 100, which is software that implements each function, and the control program and various data are recorded so as to be readable by a computer (or CPU).
  • a ROM (Read Only Memory) or a storage device (these are called “recording media”), a RAM (Random Access Memory) for expanding the control program, and the like are provided.
  • the computer reads the control program from the recording medium and executes it, thereby achieving the object of the present invention.
  • a “non-temporary tangible medium” such as a tape, a disk, a card, a semiconductor memory, a programmable logic circuit, or the like can be used.
  • the control program may be supplied to the computer via any transmission medium (such as a communication network or a broadcast wave) that can transmit the control program.
  • the present invention can also be realized in the form of a data signal embedded in a carrier wave, in which the control program is embodied by electronic transmission.
  • the data analysis device control program is a data analysis device control program capable of extracting data related to a predetermined case from a plurality of data acquired from around the vehicle. Therefore, the data analysis apparatus is allowed to realize a relationship evaluation function and a data notification function.
  • the relationship evaluation function and the data notification function can be realized by the relationship evaluation unit 16 and the data notification unit 18 described above, respectively. Details are as described above.
  • the above control program is implemented using, for example, a script language such as Python, ActionScript, JavaScript (registered trademark), an object-oriented programming language such as Objective-C, Java (registered trademark), or a markup language such as HTML5. it can.
  • a script language such as Python, ActionScript, JavaScript (registered trademark), an object-oriented programming language such as Objective-C, Java (registered trademark), or a markup language such as HTML5. it can.
  • the element evaluation unit determines a transmission information amount representing a dependency relationship between the data element and a result determined by the driver with respect to already determined data including the data element. As one of the criteria, the data element can be evaluated.
  • a data analysis apparatus acquires digital information including data, driver information, and access history information, specifies a specific driver from the driver information, and based on the access history information regarding the specified specific driver Only the data accessed by a specific driver is extracted, and additional information indicating whether or not a predetermined file included in the extracted data is related to a predetermined case is set. To output a predetermined file related to the predetermined case.
  • a data analysis apparatus acquires digital information including data and driver information, and sets driver identification information indicating which driver among the drivers included in the driver information is related. , Specifying a driver, searching for a predetermined file in which driver identification information corresponding to the specified driver is set, and indicating whether or not the searched predetermined file is related to a predetermined case Information is set, and a predetermined file related to a predetermined case is output based on the incidental information.
  • a data analysis apparatus includes: (1a) a classification code A, (1b) a data element included in data provided with a classification code A, and (1c) a classification code A and a data element. Is stored in the related data element database, and (2a) the classification code B and (2b) the related data element having a high appearance frequency in the data to which the classification code B is assigned, ( 2c) Related data element correspondence information indicating the correspondence between the classification code B and the related data element is stored, and based on the data element correspondence information of (1c), data including the data element of (1b) is stored.
  • the data including the related data element (2b) is extracted from the data to which the classification code A is assigned and the classification code A is not given, and the evaluation value / number of the related data element is extracted.
  • the classification code B is given to the data whose score exceeds a certain value, and the classification code B is not given to the data.
  • the application of the classification code C is accepted from the driver.
  • a data analysis apparatus receives a classification code input from a driver and classifies data for each classification code in order to give a classification code indicating relevance to a predetermined case to data. Analyzing and selecting common data elements in the sorted data, searching the selected data elements from the data, and using the search results and the results of analyzing the data elements, the classification code and data The score which shows the relevance to is calculated, and a classification code is assigned to the data based on the calculated score.
  • a data analysis apparatus registers a data element for determining whether a driver is related to a predetermined case in a database, searches the data for a data element registered in the database, and performs a search
  • the sentence including the selected data element is extracted from the data, and a score indicating the degree of relevance with a predetermined case is calculated from the feature amount extracted from the extracted sentence, and the degree of sentence emphasis is determined according to the score. Change.
  • a data analysis apparatus records a result of a relevance determination with a predetermined case performed by a driver or a progress speed of the relevance determination as performance information, and generates prediction information related to the result or the progress speed Then, the performance information and the prediction information are compared, and based on the comparison result, an icon for presenting an evaluation for the relevance judgment of the driver is generated.
  • a data analysis apparatus receives input from a driver for result information indicating a relationship between data and a predetermined case, and from the characteristics of data elements that appear in common in the data, An evaluation value is calculated for each result information, a data element is selected based on the evaluation value, a data score is calculated from the selected data element and its evaluation value, and a recall is calculated based on the score.
  • the data analysis device displays data for a driver, and provides identification information (based on a determination as to whether or not the driver is related to a predetermined case with respect to data to be reviewed) Tag), the feature quantity of the target data that received the tag is compared with the feature quantity of the data, the score of the data corresponding to the predetermined tag is updated based on the comparison result, and based on the updated score To control the display order of the displayed data.
  • the data analysis apparatus When the source code is updated, the data analysis apparatus according to one aspect of the present invention records the updated source code, creates an executable file from the recorded source code, and verifies the executable file The verification result is executed and the server receives the delivery of the verification result.
  • a data analysis apparatus displays data for a driver to determine the relevance to a predetermined case, and a classification button for causing the driver to select a classification condition for classifying the data.
  • the information regarding the classification button selected by is received as selection information, the data is classified based on the result of analyzing the data based on the selection information, and the data is displayed based on the classification result.
  • the data analysis apparatus confirms the incidental information of the audio / image data, classifies the audio / image data based on the incidental information, and includes the classified audio / image data. Are extracted, the similarity is analyzed based on the extracted elements, and integrated and analyzed based on the similarity.
  • the data analysis apparatus extracts a password-protected file protected by a password, and uses the dictionary file in which candidate words that are password candidates are registered, Input the password-resolved file and accept the determination result of the relevance with the predetermined case performed by the driver.
  • a data analysis apparatus divides binary search target file data into a plurality of blocks, searches the block search data from binary search target files, and outputs the search results. To do.
  • a data analysis apparatus selects target digital information to be investigated, stores a combination of a plurality of words having relevance to a specific matter, and stores the selected target digital information in the selected target digital information Whether or not a combination of a plurality of words is included, and if so, based on the result of the morphological analysis, the relevance to the specific matter of the target digital information is determined, and the determination result is Correspond to target digital information.
  • the data analysis apparatus receives an input of a classification code from a driver in order to extract an image group / sound group from image information / sound information, and to assign a classification code to the image group / sound group,
  • the image group / sound group is classified for each classification code, the data elements that appear in common in the sorted image group / sound group are analyzed and selected, and the selected data element is searched from the image information / sound information, Using the search result and the result of analyzing the data element, a score is calculated, and based on the calculated score, a classification code is assigned to the image information / audio information, and the score calculation result and the classification result are displayed on the screen. Then, the number of images / sounds necessary for reconfirmation is calculated based on the relationship between the recall ratio and the standardization order.
  • a data analysis apparatus includes: (1a) a classification code A, (1b) a data element included in data provided with a classification code A, and (1c) a classification code A and a data element. Is stored in the related data element database, and (2a) the classification code B and (2b) the related data element having a high appearance frequency in the data to which the classification code B is assigned, ( 2c) Related data element correspondence information indicating the correspondence between the classification code B and the related data element is stored, and based on the data element correspondence information of (1c), data including the data element of (1b) is stored.
  • the data including the related data element (2b) is extracted from the data to which the classification code A is assigned and the classification code A is not given, and the evaluation value / number of the related data element is extracted.
  • the classification code B is given to the data whose score exceeds a certain value, and the classification code B is not given to the data.
  • the application of the classification code C is received from the driver, the data to which the classification code C is assigned is analyzed, and the classification code D is given to the data to which the classification code is not given based on the analysis result.
  • the data analysis apparatus calculates a score indicating the relevance with a predetermined case for each data.
  • the data is extracted in a predetermined order based on the calculated score, and the classification code provided by the driver based on the relevance with the predetermined case is accepted for the extracted data, and extracted based on the classification code.
  • the data is classified by classification code, and in the sorted data, the data elements that appear in common are analyzed and selected, the selected data elements are searched from the data, and the score is obtained using the search result and the analysis result. Is calculated again for each data.
  • information related to a predetermined case is stored in the basic research database, the input of the category of the predetermined case is accepted, and the target of the investigation is based on the received category
  • the survey category is determined, and the necessary information type is extracted from the survey basic database.
  • a data analysis apparatus stores an action occurrence model created based on a transmission / reception history of a message file on a network of an action subject who has performed a specific action, and stores a message file on the network of the subject Based on the transmission / reception history, profile information of the subject is created, a score indicating the compatibility between the profile information and the behavior generation model is calculated, and the possibility of occurrence of a specific behavior is determined based on the score.
  • a data analysis apparatus collects a case investigation result including a sorting work result for each case with respect to a predetermined case, registers a check model parameter for checking the predetermined case, and creates a new check
  • the survey content of the project is entered, the registered survey model parameters are searched, the survey model parameters related to the input information are extracted, and the survey model is output using the extracted survey model parameters.
  • a data analysis apparatus acquires driver information related to a driver, acquires updated digital information at regular intervals based on the driver information, and recording destination information related to the acquired digital information Based on the file name and metadata, the files that make up the acquired digital information are arranged in a predetermined storage location, and the status of the arranged files is the status of the driver that accessed the digital information. Create a visualized situation distribution so that it can be understood.
  • the data analysis apparatus acquires metadata associated with digital information, and sets a weighting parameter set based on the relationship between the first digital information and the metadata having a relationship with the specific matter. And update the association between the morpheme and the digital information using the weighting parameter set.
  • a data analysis apparatus receives a classification code manually assigned to target data, calculates a relevance score of the target data, and determines whether the classification code is correct based on the relevance score Then, the classification code to be assigned to the target data is determined based on the result of the correctness determination.
  • a data analysis apparatus receives an input of a category to which a predetermined case belongs, performs a survey based on the received category, creates a report for reporting the result of the survey, and a survey basic database To store information related to a given case, determine the survey category to be surveyed based on the received category, extract the necessary information type from the survey basic database, and specify the type of extracted information
  • An input of a data element that is presented to the driver and used to give a classification code corresponding to the type of information presented is received from the driver, and a classification code is automatically assigned to the data.
  • a data analysis apparatus acquires public information of a subject, analyzes the public information, outputs an external element of the subject, and is based on an action external element of the behavior subject who has performed a specific action
  • the action generation model is stored, the action factors that match the action generation model are extracted from the external elements of the subject, stored, the internal information of the subject is obtained, the internal information is analyzed, and the internal elements of the subject are output Then, the analysis target is automatically specified based on the similarity between the internal element and the action factor.
  • a data analysis apparatus acquires relevance information indicating a relevance between digital information and a specific item from a driver, and determines a relevance score determined according to the relationship between the digital information and the specific item. Is calculated for each digital information, and for each predetermined range of relevance scores, the relevance given to the digital information included in the range with respect to the total number of digital information having relevance scores included in each range A ratio of the number of information is calculated, and a plurality of sections associated with each range are displayed with the hue, brightness, or saturation changed based on the ratio.
  • the data analysis apparatus calculates a score indicating the strength of the connection between the data and the classification code in time series, detects a time-series change in the score from the calculated score, When determining the time-series change in the detected score, the degree of association between the survey item and the extracted data is determined based on the result of determining the time when the score has exceeded a predetermined reference value.
  • a data analysis apparatus has weight information associated with a plurality of data elements including a co-occurrence expression and has a relationship with a specific matter, and associates scores with digital information. Based on the score, sample digital information as a sample is extracted from the digital information, and the weighted information is updated by analyzing the extracted sample digital information.
  • the data analysis apparatus selects a category that is an index that can classify each data included in a plurality of data, and calculates a score for each category.
  • the data analysis apparatus specifies, based on the score, a phase for classifying a predetermined action by a predetermined action subject that causes a predetermined case according to progress of the predetermined action.
  • the change of the identified phase is estimated based on the temporal transition of the phase.
  • the data analysis apparatus stores a generation process model in which a predetermined action causing a predetermined case occurs for each phase classified according to the progress of the predetermined action, and stores the generated process model in the predetermined case.
  • Stores related information for each category and generation process model stores time-series information indicating the temporal order of phases, analyzes image information and audio information based on these information, and generates a predetermined action It is calculated from the result of analyzing an index indicating possibility.
  • the data analysis apparatus stores a generation process model in which a predetermined action causing a predetermined case occurs for each phase classified according to the progress of the predetermined action, and stores the generated process model in the predetermined case.
  • Stores related information for each category and generation process model stores time-series information indicating the temporal order of phases, stores relationships among multiple people related to a given case, and stores these information in these information Analyze the data based on and identify the current phase.
  • the object specifying the object of the action is identified, metadata indicating the attribute of the sound including the verb and the object, and the metadata
  • the verb and the object are associated with each other, the relationship between the sound and the case is evaluated based on the association, and the relationship between a plurality of persons related to the case is displayed.
  • a data analysis apparatus acquires communication data that is transmitted and received between a plurality of terminals and is associated with each of a plurality of persons, analyzes the content of the acquired communication data, and uses the analysis result Then, the relationship between the contents of the communication data and the predetermined case is evaluated, and based on the evaluation result, the relationship among a plurality of persons related to the predetermined case is displayed.
  • the data analysis apparatus calculates a score indicating the strength with which data included in a data group is associated with a classification code indicating the degree of association between the data group and a predetermined case, and the calculated score In response, the score is reported to the driver, and a survey report is output according to the survey type of a predetermined case.
  • a data analysis apparatus refers to a database that stores confidential information in association with confidentiality, calculates a leakage level indicating a risk of leakage of confidential information due to access to the outside of the network, and It is determined whether the density or the leakage level satisfies a criterion for leaking confidential information. If it is determined that the information is satisfied, the subject of leakage is specified.
  • the data analysis apparatus extracts a difference between a series A and a series B by comparing a series A of behaviors included in a certain period with a series B of behaviors included in the most recent period. Then, it is determined whether or not the extracted difference has reached a standard that suggests that the risk of leakage of confidential information has increased. If it is determined that the difference has been reached, the risk is notified to the administrator.
  • a data analysis apparatus generates a data element vector indicating whether or not a predetermined data element is included in a sentence included in data (for example, speech) for each sentence, and generates the data element vector. Then, by multiplying the correlation matrix indicating the correlation between the predetermined data element and the other data elements, a correlation vector is obtained for each sentence, and a score is calculated based on the sum of all the correlation vectors.
  • the data analysis apparatus learns the weighting of data elements included in the sorted data sorted by the driver as to whether or not it relates to a predetermined case, and whether or not it relates to the predetermined case. Is searched from the unsorted data that has not yet been sorted by the driver, and the data elements included in the sorted data are searched, and the weights of the searched data elements and the learned data elements are used to calculate the unsorted data and the classification code. A score that evaluates the strength of the connection is calculated.

Landscapes

  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

La présente invention fournit des données requises par un conducteur au conducteur. La présente invention est pourvue : d'une unité d'évaluation de relation qui, lorsque des données non jugées pour lesquelles il a été jugé si les données concernent ou non un cas prescrit sont récemment acquises, évalue la relation entre les données non jugées et le cas prescrit en fonction des données jugées pour lesquelles un conducteur conduisant un véhicule a jugé si lesdites données jugées concernent ou non le cas prescrit ; et d'une unité de rapport de données qui rapporte les données non jugées au conducteur selon la relation évaluée par l'unité d'évaluation de relation.
PCT/JP2014/068004 2014-07-07 2014-07-07 Dispositif d'analyse de données, procédé de commande pour dispositif d'analyse de données, et programme de commande pour dispositif d'analyse de données WO2016006021A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2014/068004 WO2016006021A1 (fr) 2014-07-07 2014-07-07 Dispositif d'analyse de données, procédé de commande pour dispositif d'analyse de données, et programme de commande pour dispositif d'analyse de données

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2014/068004 WO2016006021A1 (fr) 2014-07-07 2014-07-07 Dispositif d'analyse de données, procédé de commande pour dispositif d'analyse de données, et programme de commande pour dispositif d'analyse de données

Publications (1)

Publication Number Publication Date
WO2016006021A1 true WO2016006021A1 (fr) 2016-01-14

Family

ID=55063698

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2014/068004 WO2016006021A1 (fr) 2014-07-07 2014-07-07 Dispositif d'analyse de données, procédé de commande pour dispositif d'analyse de données, et programme de commande pour dispositif d'analyse de données

Country Status (1)

Country Link
WO (1) WO2016006021A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2019016274A (ja) * 2017-07-10 2019-01-31 三菱ロジスネクスト株式会社 荷役作業割当システム
EP4147684A1 (fr) 2021-09-10 2023-03-15 The Procter & Gamble Company Article absorbant comprenant une couche de coussin multicouche

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007219836A (ja) * 2006-02-16 2007-08-30 Mazda Motor Corp 車両用走行支援装置
JP2008137467A (ja) * 2006-11-30 2008-06-19 Fuji Heavy Ind Ltd 車両の運転支援装置
JP2009096365A (ja) * 2007-10-17 2009-05-07 Fuji Heavy Ind Ltd リスク認識システム
JP2012185561A (ja) * 2011-03-03 2012-09-27 Internatl Business Mach Corp <Ibm> 情報処理装置、自然言語解析方法、プログラムおよび記録媒体

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007219836A (ja) * 2006-02-16 2007-08-30 Mazda Motor Corp 車両用走行支援装置
JP2008137467A (ja) * 2006-11-30 2008-06-19 Fuji Heavy Ind Ltd 車両の運転支援装置
JP2009096365A (ja) * 2007-10-17 2009-05-07 Fuji Heavy Ind Ltd リスク認識システム
JP2012185561A (ja) * 2011-03-03 2012-09-27 Internatl Business Mach Corp <Ibm> 情報処理装置、自然言語解析方法、プログラムおよび記録媒体

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2019016274A (ja) * 2017-07-10 2019-01-31 三菱ロジスネクスト株式会社 荷役作業割当システム
EP4147684A1 (fr) 2021-09-10 2023-03-15 The Procter & Gamble Company Article absorbant comprenant une couche de coussin multicouche
WO2023039401A1 (fr) 2021-09-10 2023-03-16 The Procter & Gamble Company Article absorbant comprenant une couche de coussin multicouche

Similar Documents

Publication Publication Date Title
Pedreschi et al. Meaningful explanations of black box AI decision systems
CN111971698A (zh) 在神经网络中使用梯度来检测后门
JP7032623B2 (ja) 推論装置、推論方法及びプログラム
Ghazizadeh et al. Text mining to decipher free-response consumer complaints: Insights from the NHTSA vehicle owner’s complaint database
Hajializadeh Deep learning-based indirect bridge damage identification system
EP3940487A1 (fr) Estimation de la probabilité de collision avec un niveau de gravité croissant pour véhicules autonomes
US20170358045A1 (en) Data analysis system, data analysis method, and data analysis program
US20140365403A1 (en) Guided event prediction
JP5905651B1 (ja) 実績評価装置、実績評価装置の制御方法、および実績評価装置の制御プログラム
Das et al. Evaluating lateral interactions of motorized two-wheelers using multi-gene symbolic genetic programming
Sibai AI crimes: a classification
CN110533094B (zh) 一种用于驾驶员的评价方法和系统
Chauhan et al. Challenges in rear-end conflict-based safety assessment of highly disordered traffic conditions
Završnik Big data: What is it and why does it matter for crime and social control?
WO2016006021A1 (fr) Dispositif d&#39;analyse de données, procédé de commande pour dispositif d&#39;analyse de données, et programme de commande pour dispositif d&#39;analyse de données
Gerrits Soul of a new machine: Self-learning algorithms in public administration
Huang et al. Using conditional generative adversarial nets and heat maps with simulation-accelerated training to predict the spatiotemporal impacts of highway incidents
CN112241470A (zh) 一种视频分类方法及系统
Akbal et al. Vehicle interior sound classification based on local quintet magnitude pattern and iterative neighborhood component analysis
WO2019187107A1 (fr) Dispositif de traitement d&#39;informations, procédé de commande et programme
Taamneh et al. A machine learning approach for building an adaptive, real-time decision support system for emergency response to road traffic injuries
CN114612246A (zh) 对象集合识别方法、装置、计算机设备及存储介质
Miao et al. Vehicle reidentification in a connected vehicle environment using machine learning algorithms
Török Statistical Analysis of a Multi-Criteria Assessment of Intelligent Traffic Systems for the Improvement of Road Safety
CN111563522A (zh) 用于识别图像中的干扰的方法和设备

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: 14896964

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

NENP Non-entry into the national phase

Ref country code: JP

122 Ep: pct application non-entry in european phase

Ref document number: 14896964

Country of ref document: EP

Kind code of ref document: A1