WO2020026328A1 - Dispositif de traitement d'informations, procédé de commande et programme - Google Patents

Dispositif de traitement d'informations, procédé de commande et programme Download PDF

Info

Publication number
WO2020026328A1
WO2020026328A1 PCT/JP2018/028566 JP2018028566W WO2020026328A1 WO 2020026328 A1 WO2020026328 A1 WO 2020026328A1 JP 2018028566 W JP2018028566 W JP 2018028566W WO 2020026328 A1 WO2020026328 A1 WO 2020026328A1
Authority
WO
WIPO (PCT)
Prior art keywords
feature
gas
information
target gas
odor
Prior art date
Application number
PCT/JP2018/028566
Other languages
English (en)
Japanese (ja)
Inventor
鈴木　亮太
江藤　力
Original Assignee
日本電気株式会社
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 日本電気株式会社 filed Critical 日本電気株式会社
Priority to JP2020533926A priority Critical patent/JPWO2020026328A1/ja
Priority to PCT/JP2018/028566 priority patent/WO2020026328A1/fr
Publication of WO2020026328A1 publication Critical patent/WO2020026328A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N19/00—Investigating materials by mechanical methods
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
    • G01N27/02—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance
    • G01N27/04—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance by investigating resistance
    • G01N27/12—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance by investigating resistance of a solid body in dependence upon absorption of a fluid; of a solid body in dependence upon reaction with a fluid, for detecting components in the fluid
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N5/00—Analysing materials by weighing, e.g. weighing small particles separated from a gas or liquid
    • G01N5/02—Analysing materials by weighing, e.g. weighing small particles separated from a gas or liquid by absorbing or adsorbing components of a material and determining change of weight of the adsorbent, e.g. determining moisture content

Definitions

  • the present invention relates to the analysis of gas characteristics.
  • Patent Literature 1 discloses a technique for determining the type of a sample gas using a signal (time-series data of a detection value) obtained by measuring the sample gas with a nanomechanical sensor. Specifically, the diffusion time constant of the sample gas with respect to the receptor of the sensor is determined by the combination of the type of the receptor and the type of the sample gas, and therefore, based on the diffusion time constant obtained from the signal and the type of the receptor. It is disclosed that the type of the sample gas can be determined.
  • Patent Document 1 assumes that the sample gas contains only one type of molecule, and does not assume that a sample gas in which a plurality of types of molecules are mixed is handled.
  • the present invention has been made in view of the above problems, and has as its object to provide a technique for specifying the type or component of a gas in which a plurality of types of molecules are mixed.
  • a first information processing apparatus includes: 1) a feature amount acquiring unit for acquiring a feature amount of a target gas; and 2) a odor information in which a label of an odor is associated with a feature amount of a gas generating the odor. And a label specifying unit that specifies odor information similar to the characteristic amount of the target gas and specifies the odor label indicated by the specified odor information as the odor label of the target gas.
  • the characteristic amount of the gas indicates the magnitude of each of the plurality of characteristic constants with respect to the time-series data of the detection value obtained from the sensor that sensed the gas.
  • the detection value of the sensor changes according to the attachment and detachment of molecules contained in the gas.
  • the characteristic constant is a time constant or a rate constant relating to the magnitude of a temporal change in the amount of molecules attached to the sensor.
  • a second information processing apparatus includes: 1) a feature amount acquisition unit for acquiring a feature amount of a target gas; and 2) a unit in which an identifier of a unit component is associated with a feature amount of a gas including only the unit component.
  • a component specifying unit that specifies one or more unit components contained in the target gas using the component information and the characteristic amount of the target gas.
  • the characteristic amount of the gas indicates the magnitude of each of the plurality of characteristic constants with respect to the time-series data of the detection value obtained from the sensor that sensed the gas.
  • the detection value of the sensor changes according to the attachment and detachment of molecules contained in the gas.
  • the characteristic constant is a time constant or a rate constant relating to the magnitude of a temporal change in the amount of molecules attached to the sensor.
  • the first control method of the present invention is executed by a computer.
  • the control method includes: 1) a characteristic amount obtaining step of obtaining a characteristic amount of the target gas; and 2) a characteristic of the target gas from the odor information in which the label of the odor is associated with the characteristic amount of the gas generating the odor.
  • the characteristic amount of the gas indicates the magnitude of each of the plurality of characteristic constants with respect to the time-series data of the detection value obtained from the sensor that sensed the gas.
  • the detection value of the sensor changes according to the attachment and detachment of molecules contained in the gas.
  • the characteristic constant is a time constant or a rate constant relating to the magnitude of a temporal change in the amount of molecules attached to the sensor.
  • the second control method of the present invention is executed by a computer.
  • the control method includes: 1) a feature amount obtaining step of obtaining a feature amount of a target gas; 2) unit component information in which an identifier of a unit component is associated with a feature amount of a gas including only the unit component; A component specifying step of specifying one or more unit components contained in the target gas using the characteristic amount of the target gas.
  • the characteristic amount of the gas indicates the magnitude of each of the plurality of characteristic constants with respect to the time-series data of the detection value obtained from the sensor that sensed the gas.
  • the detection value of the sensor changes according to the attachment and detachment of molecules contained in the gas.
  • the characteristic constant is a time constant or a rate constant relating to the magnitude of a temporal change in the amount of molecules attached to the sensor.
  • the program of the present invention causes a computer to execute each step of the control method of the present invention.
  • FIG. 2 is a diagram illustrating an outline of the information processing apparatus according to the first embodiment.
  • FIG. 3 is a diagram illustrating a sensor for obtaining a characteristic amount of gas handled by the information processing apparatus.
  • FIG. 2 is a diagram illustrating a functional configuration of the information processing apparatus according to the first embodiment.
  • FIG. 2 is a diagram illustrating a computer for realizing an information processing device.
  • 6 is a flowchart illustrating a flow of a process executed by the information processing apparatus according to the first embodiment. It is a figure which illustrates odor information in a table format. It is a figure which represents conversion of a feature-value notionally.
  • FIG. 9 is a diagram illustrating an outline of an information processing apparatus according to a second embodiment.
  • FIG. 9 is a diagram illustrating a functional configuration of an information processing apparatus according to a second embodiment.
  • 13 is a flowchart illustrating a flow of a process executed by the information processing apparatus according to the second embodiment. It is a figure which illustrates unit component information about a single kind of molecule in a table form. It is a figure showing the component of target gas with a graph.
  • each block diagram represents a configuration of a functional unit, not a configuration of a hardware unit.
  • FIG. 1 is a diagram illustrating an outline of an information processing apparatus 2000 according to the first embodiment.
  • the information processing apparatus 2000 according to the first embodiment specifies a label (hereinafter, an odor label) indicating the odor of the target gas based on the feature amount of the target gas.
  • the odor label indicates the name of the substance that generates the odor.
  • the smell label “apple” is specified.
  • the substance generating the smell is not limited to foods such as apples, but may be any substance such as a machine, a building material, a medicine, mold, burnt food, or garbage.
  • the smell label may represent an abstract concept such as a place or a situation where the smell is smelled.
  • odor labels such as "Cafe Smell”, “Pool Smell”, “Blue Smell Smell”, “Smell like Closet”, “Sweet Smell”, “Smell Smell”, or “Smell on a Rainy Day” Conceivable.
  • odor information In order to realize the identification of such an odor label, information that associates the odor label with the characteristic amount of the gas corresponding to the odor label is prepared in advance. This information is called odor information.
  • the information processing apparatus 2000 specifies a feature amount similar to the feature amount of the target gas from the feature amounts indicated in the odor information, and sets the odor label associated with the specified feature amount to the odor label of the target gas. To be specified.
  • the information processing device 2000 handles a feature newly found by the present inventor as a gas feature.
  • the characteristic amount of the gas handled by the information processing apparatus 2000 will be described.
  • FIG. 2 is a diagram exemplifying the sensor 10 for obtaining the characteristic amount of gas handled by the information processing apparatus 2000.
  • the sensor 10 has a receptor to which a molecule is attached, and a detection value changes according to attachment and detachment of the molecule at the receptor.
  • the time series data of the detection values output from the sensor 10 is referred to as time series data 14.
  • the time-series data 14 is also denoted as ⁇ Y ⁇
  • the detected value at the time ⁇ t ⁇ is denoted as ⁇ y (t) ⁇ , as necessary.
  • Y is a vector in which y (t) is enumerated.
  • the senor 10 is a membrane-type surface stress (MSS) sensor.
  • the MSS sensor has, as a receptor, a functional film to which a molecule adheres, and the stress generated in a support member of the functional film changes due to the attachment and detachment of the molecule to and from the functional film.
  • the MSS sensor outputs a detection value based on the change in the stress.
  • the sensor 10 is not limited to the MSS ⁇ sensor, and the physical quantity related to the viscoelasticity and dynamic characteristics (mass, moment of inertia, etc.) of the members of the sensor 10 that occur in response to the attachment and detachment of the molecule to and from the receptor. Any sensor that outputs a detection value based on the change may be used, and various types of sensors such as a cantilever type, a film type, an optical type, a piezo, and a vibration response can be employed.
  • the sensing by the sensor 10 is modeled as follows. (1) The sensor 10 is exposed to a gas containing K kinds of molecules. (2) The concentration of each molecule k contained in the gas is constant ⁇ k. (3) The sensor 10 can adsorb a total of N molecules. (4) The number of molecules k attached to the sensor 10 at time t is nk (t).
  • the time change of the number nk (t) of the molecules k attached to the sensor 10 can be formulated as follows.
  • the first and second terms on the right-hand side of the equation (1) are the increasing amount of the molecule ⁇ k ⁇ per unit time (the number of molecules ⁇ k ⁇ newly attached to the sensor 10) and the decreasing amount (the molecule ⁇ k ⁇ detached from the sensor 10). Number).
  • ⁇ k and ⁇ k are a rate constant representing the rate at which the molecule ⁇ k ⁇ adheres to the sensor 10 and a rate constant representing the rate at which the molecule ⁇ k ⁇ separates from the sensor 10, respectively.
  • the concentration ⁇ k is constant
  • the number nk (t) of the numerator k at time t can be formulated from the above equation (1) as follows.
  • nk (t) is expressed as follows.
  • the detection value of the sensor 10 is determined by the stress applied to the sensor 10 by the molecules contained in the gas. Then, it is considered that the stress acting on the sensor 10 by a plurality of molecules can be represented by a linear sum of the stress acting on each molecule. However, it is considered that the stress generated by the molecule differs depending on the type of the molecule. That is, it can be said that the contribution of the molecule to the detection value of the sensor 10 differs depending on the type of the molecule.
  • the detection value y (t) of the sensor 10 can be formulated as follows.
  • both ⁇ k and ⁇ k represent the contribution of the numerator k to the detection value of the sensor 10.
  • the purge gas is a gas used when removing the gas to be measured from the sensor 10.
  • the time-series data 14 obtained from the sensor 10 that sensed the gas can be decomposed as in the above equation (4), the types of molecules contained in the gas and the ratio of each type of molecule contained in the gas Can be grasped. That is, by the decomposition shown in Expression (4), data representing the characteristics of the gas (that is, the characteristic amount of the gas) is obtained.
  • the information processing apparatus 2000 handles a feature amount obtained by decomposing the time-series data 14 as shown in the following equation (5).
  • ⁇ i ⁇ is a constant called a feature constant.
  • ⁇ I ⁇ is a contribution value representing the contribution of the characteristic constant ⁇ i ⁇ to the detection value of the sensor 10.
  • equation (5) can be expressed as follows.
  • the association between the set of feature constants ⁇ ⁇ and the set of contribution values ⁇ is represented, for example, by m feature matrix ⁇ F ⁇ having two rows and two columns (m is the number of each of the feature constant and the contribution value).
  • F ( ⁇ T, ⁇ T) ⁇ .
  • the vector representing the set of characteristic constants may be omitted in the characteristic amount of the gas.
  • the characteristic amount of the gas is represented by a set of contribution values.
  • the information processing device 2000 acquires a feature amount (a set of feature constants may be omitted) that associates a set of feature constants with a set of contribution values for the target gas.
  • the feature quantity indicated by the odor information is obtained by associating a set of feature constants with a set of contribution values for time-series data obtained by sensing the gas specified by the odor label indicated by the odor information with the sensor 10.
  • Information The information processing apparatus 2000 compares such characteristic amounts with each other to specify odor information indicating a characteristic amount similar to the characteristic amount of the target gas. Then, the information processing device 2000 specifies the odor label indicated by the specified odor information as the odor label of the target gas.
  • the information processing apparatus 2000 specifies the odor label of the target gas using the feature amount in which the feature constant vector and the contribution vector obtained for the time-series data of the detected gas value are associated with each other. As described above, since this characteristic amount changes depending on the molecules contained in the gas and the mixing ratio thereof, the gas can be distinguished with high accuracy. Therefore, by using such a characteristic amount, according to the information processing apparatus 2000 of the present embodiment, the odor label of the target gas can be accurately specified.
  • FIG. 1 The above description with reference to FIG. 1 is an example for facilitating understanding of the information processing device 2000, and does not limit the functions of the information processing device 2000.
  • the information processing apparatus 2000 of the present embodiment will be described in more detail.
  • FIG. 3 is a diagram illustrating a functional configuration of the information processing apparatus 2000 according to the first embodiment.
  • the information processing apparatus 2000 according to the first embodiment includes a feature amount acquiring unit 2020 and a label specifying unit 2040.
  • the feature amount acquisition unit 2020 acquires the feature amount of the target gas.
  • the label specifying unit 2040 extracts, from the plurality of pieces of odor information, odor information indicating a characteristic amount similar to the characteristic amount of the target gas. Further, the label specifying unit 2040 specifies the odor label indicated in the extracted odor information as the odor label of the target gas.
  • Each functional component of the information processing apparatus 2000 may be implemented by hardware (eg, a hard-wired electronic circuit or the like) that implements each functional component, or a combination of hardware and software (eg: Electronic circuit and a program for controlling the same).
  • hardware eg, a hard-wired electronic circuit or the like
  • software eg: Electronic circuit and a program for controlling the same.
  • FIG. 4 is a diagram illustrating a computer 1000 for realizing the information processing device 2000.
  • the computer 1000 is an arbitrary computer.
  • the computer 1000 is a stationary computer such as a personal computer (PC) or a server machine.
  • the computer 1000 is a portable computer such as a smartphone or a tablet terminal.
  • the computer 1000 may be a dedicated computer designed to realize the information processing device 2000, or may be a general-purpose computer.
  • the computer 1000 has a bus 1020, a processor 1040, a memory 1060, a storage device 1080, an input / output interface 1100, and a network interface 1120.
  • the bus 1020 is a data transmission path through which the processor 1040, the memory 1060, the storage device 1080, the input / output interface 1100, and the network interface 1120 mutually transmit and receive data.
  • a method for connecting the processors 1040 and the like to each other is not limited to a bus connection.
  • the processor 1040 is various processors such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and an FPGA (Field-Programmable Gate Array).
  • the memory 1060 is a main storage device realized using a RAM (Random Access Memory) or the like.
  • the storage device 1080 is an auxiliary storage device realized using a hard disk, an SSD (Solid State Drive), a memory card, or a ROM (Read Only Memory).
  • the input / output interface 1100 is an interface for connecting the computer 1000 and an input / output device.
  • an input device such as a keyboard and an output device such as a display device are connected to the input / output interface 1100.
  • the network interface 1120 is an interface for connecting the computer 1000 to a communication network.
  • the communication network is, for example, a LAN (Local Area Network) or a WAN (Wide Area Network).
  • the method by which the network interface 1120 connects to the communication network may be a wireless connection or a wired connection.
  • the storage device 1080 stores a program module that implements each functional component of the information processing apparatus 2000.
  • the processor 1040 realizes a function corresponding to each program module by reading out each of these program modules into the memory 1060 and executing them.
  • FIG. 5 is a flowchart illustrating a flow of a process executed by the information processing apparatus 2000 according to the first embodiment.
  • the characteristic amount acquisition unit 2020 acquires the characteristic amount of the target gas (S102).
  • the label specifying unit 2040 extracts odor information indicating a characteristic amount similar to the characteristic amount of the target gas from the plurality of odor information (S104).
  • the label specifying unit 2040 specifies the odor label indicated by the extracted odor information as the odor label of the target gas (S106).
  • the feature amount acquisition unit 2020 acquires the feature amount of the target gas.
  • the characteristic amount acquiring unit 2020 acquires the characteristic amount of the target gas by accessing a storage device in which the characteristic amount of the target gas is stored. This storage device may be provided inside the information processing device 2000 or may be provided outside the information processing device 2000.
  • the information processing apparatus 2000 may acquire the characteristic amount of the target gas by receiving the characteristic amount of the target gas transmitted from another device.
  • the “other device” is, for example, a device that calculates the feature amount of the target gas using the time-series data 14 obtained from the sensor 10 for the target gas.
  • FIG. 6 is a diagram illustrating the odor information in a table format.
  • the table in FIG. 6 is called a table 200.
  • the table 200 has two columns of an odor label 202 and a feature 204.
  • the odor label indicated by the odor label 202 is associated with the characteristic amount (in this example, the characteristic matrix F) indicated by the characteristic amount 204.
  • the characteristic amount in this example, the characteristic matrix F
  • the odor information is stored in advance in a storage device provided inside or outside the information processing device 2000.
  • the label specifying unit 2040 extracts odor information indicating a characteristic amount similar to the characteristic amount of the target gas from the plurality of odor information (S104). For example, the label specifying unit 2040 extracts odor information indicating a feature amount most similar to the feature amount of the target gas. In this case, the odor label of the target gas is uniquely specified.
  • the odor information extracted by the label specifying unit 2040 may be plural.
  • the label specifying unit 2040 extracts one or more pieces of odor information indicating a characteristic amount whose similarity to the characteristic amount of the target gas is equal to or greater than a threshold.
  • one or more odor labels having a high probability of being a label indicating the odor of the target gas are specified. That is, one or more candidates for the odor label of the target gas are specified.
  • the label specifying unit 2040 outputs that there is no odor label candidate of the target gas when there is no odor information indicating the characteristic amount whose similarity with the characteristic amount of the target gas is equal to or more than the threshold value in the table 200. You may. In this case, the target gas can be interpreted as a new smell not registered in the table 200.
  • the label specifying unit 2040 calculates the similarity of the target gas by comparing the characteristic amount of the target gas with the characteristic amount indicated by the odor information.
  • a method of calculating the similarity will be exemplified.
  • the label specifying unit 2040 extracts the odor information indicating the feature amount having the shortest distance from the feature amount of the target gas.
  • a threshold value is set for the distance from the feature amount of the target gas.
  • the label specifying unit 2040 extracts each odor information indicating a feature amount whose distance from the feature amount of the target gas is equal to or less than a threshold value.
  • the label specifying unit 2040 converts one or both of the characteristic amount of the target gas and the characteristic amount of the odor information such that a set of these characteristic constants is common. Such conversion enables similarity determination using the distance between the feature amounts described above. Therefore, the label specifying unit 2040 performs similarity determination using the distance between the feature amounts using the converted feature amount, thereby extracting odor information indicating a feature amount similar to the feature amount of the target gas.
  • FIG. 7 is a diagram conceptually showing the conversion of the feature amount.
  • ⁇ ai ⁇ associated with the feature constant ⁇ ai in the feature quantity before conversion is associated with ⁇ bj that satisfies ⁇ bj ⁇ ⁇ ai ⁇ _ (bj + 1) ⁇ in the feature quantity after conversion.
  • the conversion of the characteristic amount may be performed for one of the characteristic amount of the target gas and the characteristic amount of the odor information, or may be performed for both of them.
  • the label specifying unit 2040 may convert the characteristic amount of the target gas to match the set of characteristic constants of the characteristic amount of the odor information, or may convert the characteristic amount of the target gas into a set of characteristic constants of the target gas.
  • the feature amount of the odor information may be converted so as to match.
  • the set of characteristic constants is not common to a plurality of pieces of odor information, it is preferable to convert the characteristic amount of the odor information to match the set of characteristic constants of the characteristic amount of the target gas. This is because the similarity determination is performed after setting the set of feature constants common to all feature amounts.
  • ⁇ When converting both the characteristic amount of the target gas and the characteristic amount of the odor information, a common set ⁇ c ⁇ of characteristic constants is prepared in advance. Then, the label specifying unit 2040 converts both the characteristic amount of the target gas and the characteristic amount of the odor information so that the set of characteristic constants becomes ⁇ c ⁇ .
  • the method of similarity determination when the set of feature constants is not common is not limited to the method of performing similarity determination after converting feature values as described above.
  • the label specifying unit 2040 treats the two feature values to be compared as a probability distribution, and determines the similarity of these feature values as KL (Kullback- Leibler) Calculate divergence.
  • KL divergence is an index value indicating the degree of similarity between probability distributions, and becomes smaller as the difference between two probability distributions to be compared is smaller. Therefore, the label specifying unit 2040 treats the smaller the ⁇ KL ⁇ divergence calculated between the characteristic amount of the target gas and the characteristic amount indicated by the odor information, the more similar these are.
  • a kernel method can be used.
  • the label specifying unit 2040 calculates KL divergence by treating the feature value F as a probability distribution represented by the following equation.
  • g is a kernel function
  • C is a normalization constant of the probability distribution.
  • the label specifying unit 2040 may use, for example, Wasserstein metric as the similarity between the two feature amounts to be compared.
  • the Wasserstein metric is a metric that represents the minimum cost of moving one distribution to the other, and the smaller the difference between the two distributions to be compared, the smaller the value. Therefore, the label identifying unit 2040 treats the smaller the Wasserstein metric calculated between the characteristic amount of the target gas and the characteristic amount indicated by the odor information, as the similarity thereof. Note that the Wasserstein metric is applicable even when the contribution value ⁇ includes a negative value.
  • the information processing device 2000 outputs information indicating the odor label of the target gas (hereinafter, output information).
  • the output information is text data representing an odor label of the target gas.
  • the output information may include odor information indicating the specified odor label.
  • the output information may indicate the characteristic amount of the target gas and the characteristic amount indicated by the odor information by graphical information such as a graph or a table.
  • FIG. 8 is a view showing the characteristic amount of the target gas and the characteristic amount indicated by the odor information in a graph.
  • the solid line indicates the characteristic amount of the target gas.
  • the dotted line indicates the feature amount indicated by the odor information having the label “smell A”.
  • the information processing apparatus 2000 when a plurality of odor labels are specified, it is preferable that the information processing apparatus 2000 output these odor labels in order.
  • the information processing apparatus 2000 outputs the odor label in the order of the degree of similarity to the feature amount of the target gas (in order of decreasing distance).
  • the information processing device 2000 stores the output information in an arbitrary storage device.
  • the information processing apparatus 2000 causes the display device to display the output information.
  • the information processing device 2000 may transmit the output information to a device other than the information processing device 2000.
  • FIG. 9 is a diagram illustrating an outline of an information processing apparatus 2000 according to the second embodiment.
  • the information processing apparatus 2000 according to the second embodiment specifies the component of the target gas based on the feature amount of the target gas.
  • the specification of the component of the target gas includes at least specifying one or more unit components contained in the target gas.
  • the specification of the component of the target gas may include specifying the concentration of each unit component contained in the target gas and specifying the mixing ratio (relative concentration) of each unit component.
  • the unit component is, for example, a single type of molecule.
  • the information processing apparatus 2000 estimates one or more types of molecules contained in the target gas and estimates the concentration ratio of a plurality of molecules based on the feature amount of the target gas. Unit components other than a single type of molecule will be described later.
  • the unit component is a combination of molecules that generate a specific odor.
  • the specific odor is the odor specified by the odor label described in the first embodiment.
  • a combination of molecules that produce an apple odor that is, a combination of molecules contained in gas produced from an apple
  • the unit component information is information in which an identifier of the unit component is associated with a feature amount of the unit component.
  • the feature amount of a unit component is a feature amount calculated for time-series data obtained by sensing a gas containing only the unit component with the sensor 10.
  • the characteristic amount of the target gas and the characteristic amount of each unit component are represented by a vector expression
  • the characteristic amount of the target gas can be represented by a linear sum of the characteristic amounts of the unit components included in the target gas.
  • the characteristic amount of the target gas can be expressed as follows.
  • ⁇ i is the feature vector of the unit component i
  • ai is the concentration of the unit component i in the target gas.
  • the information processing apparatus 2000 uses the unit component information to decompose the feature vector ⁇ g of the target gas into a linear sum of the feature vectors ⁇ i of one or more unit components. By doing so, the information processing apparatus 2000 specifies one or more unit components contained in the target gas.
  • various existing methods can be used as a method of decomposing a certain vector into a linear sum of known vectors (here, each feature amount vector indicated in unit component information). .
  • a least-squares method with a non-negative constraint represented by the following objective function can be used.
  • the mixing ratio of the unit components can be represented by the ratio of these concentrations.
  • the concentration of the unit component is a value represented by (concentration of gas in air) ⁇ (concentration ratio of unit component in gas). That is, it means the relative concentration of the unit component with respect to the concentration of the gas in the air.
  • the concentration of the unit component as used herein can be regarded as a ratio of the partial pressure of the unit component to the air pressure.
  • each feature amount vector may be normalized in advance.
  • the characteristic amount ⁇ g of the target gas is decomposed as follows.
  • the set of feature constants is not the same between the feature amount of the target gas and the feature amount of the unit component, the set of feature constants is made to match by the above-described method, and then the feature amount of the target gas is changed to the unit component. Decompose into linear sum of features.
  • the information processing apparatus 2000 specifies one or more unit components included in the target gas using a feature amount determined based on the contribution of each feature constant to the time-series data of the detected gas value.
  • this characteristic amount is a characteristic amount that changes depending on the molecules contained in the gas and the mixing ratio thereof, the characteristic amount is compared with the characteristic amount of each unit component to obtain the unit component contained in the target gas. And its mixing ratio can be specified with high accuracy.
  • FIG. 10 is a diagram illustrating a functional configuration of the information processing apparatus 2000 according to the second embodiment.
  • the information processing apparatus 2000 according to the second embodiment includes a feature amount acquisition unit 2020 and a component identification unit 2060.
  • the feature amount acquisition unit 2020 is as described in the first embodiment.
  • the component specifying unit 2060 specifies one or more unit components included in the target gas by using the unit component information and the characteristic amount of the target gas.
  • the hardware configuration of a computer that implements the information processing apparatus 2000 according to the second embodiment is represented by, for example, FIG. 4 as in the first embodiment.
  • the storage device 1080 of the computer 1000 that implements the information processing apparatus 2000 of the present embodiment stores a program module that implements the functions of the information processing apparatus 2000 of the present embodiment.
  • FIG. 11 is a flowchart illustrating a flow of a process executed by the information processing apparatus 2000 according to the second embodiment.
  • the feature amount acquisition unit 2020 acquires the feature amount vector ⁇ g of the target gas (S402).
  • the component specifying unit 2060 specifies one or more unit components contained in the target gas using the acquired feature amount vector ⁇ g and the unit component information (S404).
  • the unit component information is information in which the identifier of the unit component is associated with the feature amount of the unit component.
  • the unit component is a single type of molecule.
  • the unit component information associates an identifier of a molecule with a feature amount of the molecule.
  • the identifier of a molecule is the name or chemical formula of the molecule.
  • the feature amount of a molecule is a feature amount (for example, feature matrix F) obtained by decomposing time-series data obtained by sensing a gas containing only the molecule with the sensor 10. It is assumed that the unit component information is stored in a storage device provided inside or outside the information processing device 2000 in advance.
  • FIG. 12 is a diagram exemplifying unit component information on a single type of molecule in a table format.
  • the table in FIG. 12 is called a table 300.
  • the table 300 has two columns of a molecule identifier 302 and a feature amount 304.
  • the feature amount indicated by the feature amount 304 is associated with the molecule identifier (unit component identifier) indicated by the molecule identifier 302.
  • the unit component is a combination of molecules that generate a specific odor.
  • “Specific odor” corresponds to the odor represented by the odor label described in the first embodiment.
  • the combination of molecules that generate a specific odor corresponds to the combination of molecules contained in the gas specified by the odor label.
  • “combination of molecules” not only which molecules are contained, but also the mixing ratio of those molecules is specified.
  • the identifier of the unit component indicated by the unit component information is an odor label.
  • the characteristic amount of the unit component is a characteristic amount of the gas specified by the odor label. That is, the unit component information corresponds to the odor information in the first embodiment (see FIG. 6).
  • the information processing device 2000 outputs information indicating a component of the target gas (hereinafter, second output information).
  • the second output information is text data indicating the identifier of each unit component contained in the target gas and their concentration and mixing ratio.
  • the second output information may be graphical information in which the identifiers of the unit components contained in the target gas and their concentrations and mixing ratios are expressed in a table or a graph.
  • FIG. 13 is a graph showing the components of the target gas in a graph.
  • the horizontal axis indicates the name of each molecule contained in the target gas
  • the vertical axis indicates the concentration of each molecule. Specifically, it indicates that the target gas contains molecules B, C, E, and G, and the concentrations thereof.
  • the unit components are sorted in descending order of density.
  • a feature amount obtaining unit that obtains a feature amount of the target gas; From the odor information in which the odor label and the characteristic amount of the gas generating the odor are associated, the odor information similar to the characteristic amount of the target gas is specified, and the odor label indicated in the specified odor information is specified.
  • a label specifying unit for specifying as a label of the odor of the target gas The feature value of the gas represents the magnitude of each of a plurality of feature constants with respect to the time-series data of the detection value obtained from the sensor that sensed the gas, The detection value of the sensor changes according to the attachment and detachment of molecules contained in the gas,
  • the information processing apparatus wherein the feature constant is a time constant or a rate constant relating to a magnitude of a temporal change in an amount of a molecule attached to the sensor.
  • the label identification unit calculates the similarity between the distribution of the contribution value represented by the characteristic amount of the target gas and the distribution of the contribution value represented by the characteristic amount indicated by each of the odor information, and uses the calculated similarity.
  • the gas feature amount is a feature vector in which contribution values indicating the magnitude of contribution of each feature constant are listed,
  • the label specifying unit calculates the distance between the feature vector of the target gas and the feature vector indicated by each of the odor information, and displays the label indicated by the odor information having the calculated distance that is the minimum, the label of the target gas. 1. Identify as odor label An information processing apparatus according to claim 1. 4.
  • the gas feature quantity is information that associates a set of feature constants with a set of contribution values of each feature constant
  • the label identification unit Both or one of the feature quantity of the target gas and the feature quantity indicated by the odor information is set so that the set of feature constants indicated by the feature quantity of the target gas and the set of feature constants indicated by the odor information become the same. Converted, After the conversion, a distance between a feature vector indicating a set of contribution values indicated by the feature amount of the target gas and a feature vector indicating a set of contribution values indicated by each of the odor information is calculated, and the calculated distance is calculated. 1.
  • the label indicated by the minimum odor information is specified as the odor label of the target gas.
  • the gas feature quantity is information that associates a set of feature constants with a set of contribution values of each feature constant
  • the label identification unit A Kullback-Leibler (KL) divergence or Wasserstein metric is calculated between the distribution of the contribution value represented by the characteristic amount of the target gas and the distribution of the contribution value represented by the characteristic amount indicated by each of the odor information. 1. Specify the label indicated by the odor information having the minimum KL divergence or Wasserstein weighing as the odor label of the target gas.
  • KL Kullback-Leibler
  • a feature amount obtaining unit that obtains a feature amount of the target gas; Identifying one or more unit components contained in the target gas using unit component information in which an identifier of a unit component is associated with a feature amount of a gas including only the unit component and the feature amount of the target gas And a component specifying part,
  • the feature value of the gas represents the magnitude of each of a plurality of feature constants with respect to the time-series data of the detection value obtained from the sensor that sensed the gas, The detection value of the sensor changes according to the attachment and detachment of molecules contained in the gas,
  • the information processing apparatus wherein the feature constant is a time constant or a rate constant relating to a magnitude of a temporal change in an amount of a molecule attached to the sensor.
  • the component specifying unit decomposes the characteristic amount of the target gas into one or more characteristic amounts indicated in the unit component information, and converts a unit component corresponding to each of the characteristic amounts obtained by the decomposition into the target component. 5. specified as the unit component contained in the gas; An information processing apparatus according to claim 1. 8.
  • the characteristic amount of the gas indicates a characteristic vector in which contribution values indicating the magnitude of contribution of each characteristic constant are listed,
  • the component identification unit decomposes the feature vector of the target gas into a linear sum of the feature vectors indicated by each of the one or more unit component information, and generates a unit component corresponding to the feature vector forming the linear sum, 6. Identify as the unit component contained in the target gas; An information processing apparatus according to claim 1. 9.
  • the gas feature quantity is information that associates a set of feature constants with a set of contribution values of each feature constant
  • the component identification unit Both the characteristic amount of the target gas and the characteristic amount indicated by the unit component information or the characteristic amount indicated by the unit component information so that the set of characteristic constants indicated by the characteristic amount of the target gas and the set of characteristic constants indicated by the unit component information become the same. Convert one, After the conversion, a feature vector representing a set of contribution values indicated by the feature values of the target gas is decomposed into a linear sum of feature vectors representing a set of contribution values indicated by the feature values of each of one or more unit components, 6.
  • the component specifying unit specifies a mixture ratio of each of the unit components in the target gas based on a coefficient of a feature vector of each of the unit components in the linear sum.
  • the unit component is a single type of molecule or a combination of molecules constituting a gas that generates a specific odor.
  • a control method executed by a computer A feature value obtaining step of obtaining a feature value of the target gas; From the odor information in which the odor label and the characteristic amount of the gas generating the odor are associated, the odor information similar to the characteristic amount of the target gas is specified, and the odor label indicated in the specified odor information is specified.
  • a label specifying step of specifying as a label of the odor of the target gas The feature value of the gas represents the magnitude of each of a plurality of feature constants with respect to the time-series data of the detection value obtained from the sensor that sensed the gas, The detection value of the sensor changes according to the attachment and detachment of molecules contained in the gas,
  • the control method wherein the characteristic constant is a time constant or a rate constant relating to a magnitude of a temporal change in an amount of a molecule attached to the sensor. 13.
  • the control method described in 1. 14 The gas feature amount is a feature vector in which contribution values indicating the magnitude of contribution of each feature constant are listed, In the label specifying step, the distance between the feature vector of the target gas and the feature vector indicated by each of the odor information is calculated, and the label indicated by the odor information with the calculated distance being the minimum is set to the label of the target gas. 12. Identify as odor label; The control method described in 1. 15.
  • the gas feature quantity is information that associates a set of feature constants with a set of contribution values of each feature constant
  • Both or one of the feature quantity of the target gas and the feature quantity indicated by the odor information is set so that the set of feature constants indicated by the feature quantity of the target gas and the set of feature constants indicated by the odor information become the same. Converted, After the conversion, a distance between a feature vector indicating a set of contribution values indicated by the feature amount of the target gas and a feature vector indicating a set of contribution values indicated by each of the odor information is calculated, and the calculated distance is calculated. 12. Identify the label indicated by the minimum odor information as the odor label of the target gas; The control method described in 1. 16.
  • the gas feature quantity is information that associates a set of feature constants with a set of contribution values of each feature constant
  • a Kullback-Leibler (KL) divergence or Wasserstein metric is calculated between the distribution of the contribution value represented by the characteristic amount of the target gas and the distribution of the contribution value represented by the characteristic amount indicated by each of the odor information. 12. Identify the label indicated by the odor information having the minimum KL divergence or Wasserstein weighing as the odor label of the target gas; The control method described in 1.
  • a control method executed by a computer A feature value obtaining step of obtaining a feature value of the target gas; Identifying one or more unit components contained in the target gas using unit component information in which an identifier of a unit component is associated with a feature amount of a gas including only the unit component and the feature amount of the target gas Component identification step,
  • the feature value of the gas represents the magnitude of each of a plurality of feature constants with respect to the time-series data of the detection value obtained from the sensor that sensed the gas,
  • the detection value of the sensor changes according to the attachment and detachment of molecules contained in the gas
  • the control method wherein the characteristic constant is a time constant or a rate constant relating to a magnitude of a temporal change in an amount of a molecule attached to the sensor.
  • the characteristic amount of the target gas is decomposed into one or more characteristic amounts indicated in the unit component information, and a unit component corresponding to each of the characteristic amounts obtained by the decomposition is converted into the target component. 17. Specified as the unit component contained in the gas; The control method described in 1. 19.
  • the characteristic amount of the gas indicates a characteristic vector in which contribution values indicating the magnitude of contribution of each characteristic constant are listed,
  • the feature vector of the target gas is decomposed into a linear sum of feature vectors indicated by each of the one or more unit component information, and a unit component corresponding to the feature vector forming the linear sum is defined as 17. specified as the unit component contained in the target gas; The control method described in 1. 20.
  • the gas feature quantity is information that associates a set of feature constants with a set of contribution values of each feature constant
  • the component identification step Both the characteristic amount of the target gas and the characteristic amount indicated by the unit component information or the characteristic amount indicated by the unit component information so that the set of characteristic constants indicated by the characteristic amount of the target gas and the set of characteristic constants indicated by the unit component information become the same. Convert one, After the conversion, a feature vector representing a set of contribution values indicated by the feature values of the target gas is decomposed into a linear sum of feature vectors representing a set of contribution values indicated by the feature values of each of one or more unit components, 17. specifying a unit component corresponding to the feature vector forming the linear sum as the unit component included in the target gas; The control method described in 1.
  • a mixing ratio of each unit component in the target gas is specified based on a coefficient of a feature vector of each unit component in the linear sum.
  • the unit component is a single type of molecule or a combination of molecules constituting a gas that generates a specific odor. To 21. The control method according to any one of the above.

Landscapes

  • Chemical & Material Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Electrochemistry (AREA)
  • Investigating Or Analyzing Materials By The Use Of Fluid Adsorption Or Reactions (AREA)

Abstract

L'invention concerne un dispositif de traitement d'informations (2000) qui acquiert des valeurs de grandeur pour un gaz étudié qui associent un ensemble de constantes caractéristiques et un ensemble de valeurs de contribution. Le dispositif de traitement d'informations (2000) spécifie des informations olfactives indiquant des valeurs de grandeur similaires aux valeurs de grandeur du gaz étudié et spécifie, en tant que marqueur olfactif pour le gaz étudié, un marqueur olfactif indiqué par les informations olfactives spécifiées. Les informations olfactives associent des marqueurs olfactifs aux valeurs de grandeur de gaz produisant les odeurs correspondant aux marqueurs olfactifs. Les valeurs de grandeur de gaz pour un gaz donné indiquent la quantité selon laquelle chacune d'une pluralité de constantes de grandeur contribue à des données chronologiques de valeurs de détection obtenues à partir d'un capteur qui a détecté ce gaz. Les valeurs de détection de capteur varient en fonction de l'adhésion et du décollement de molécules incluses dans le gaz. Les constantes de grandeur sont des constantes de temps ou des constantes de vitesse se rapportant à la quantité de variation au cours du temps de la quantité de molécules adhérant au capteur.
PCT/JP2018/028566 2018-07-31 2018-07-31 Dispositif de traitement d'informations, procédé de commande et programme WO2020026328A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2020533926A JPWO2020026328A1 (ja) 2018-07-31 2018-07-31 情報処理装置、制御方法、及びプログラム
PCT/JP2018/028566 WO2020026328A1 (fr) 2018-07-31 2018-07-31 Dispositif de traitement d'informations, procédé de commande et programme

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2018/028566 WO2020026328A1 (fr) 2018-07-31 2018-07-31 Dispositif de traitement d'informations, procédé de commande et programme

Publications (1)

Publication Number Publication Date
WO2020026328A1 true WO2020026328A1 (fr) 2020-02-06

Family

ID=69232433

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2018/028566 WO2020026328A1 (fr) 2018-07-31 2018-07-31 Dispositif de traitement d'informations, procédé de commande et programme

Country Status (2)

Country Link
JP (1) JPWO2020026328A1 (fr)
WO (1) WO2020026328A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPWO2021186532A1 (fr) * 2020-03-17 2021-09-23
US11974803B2 (en) 2020-10-12 2024-05-07 Biosense Webster (Israel) Ltd. Basket catheter with balloon

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06308008A (ja) * 1993-04-26 1994-11-04 Nippon Telegr & Teleph Corp <Ntt> 混合ガス成分判別および定量方法ならびに装置
JPH11264809A (ja) * 1998-03-17 1999-09-28 Shimadzu Corp ガス識別装置
JP2001305034A (ja) * 2000-04-26 2001-10-31 Mitsubishi Electric Corp ガス同定方法、ガス定量方法およびガス同定または定量装置
JP2011203006A (ja) * 2010-03-24 2011-10-13 Olympus Corp 検出センサ
US20160084808A1 (en) * 2014-09-20 2016-03-24 Commissariat A L'energie Atomique Et Aux Energies Alternatives Method and device for determining a composition of a gas sample processed by means of gas chromatography
US20160161433A1 (en) * 2014-05-16 2016-06-09 Massachusetts Institute Of Technology Electrospun Polymer Fibers for Gas Sensing
JP2017117438A (ja) * 2015-11-13 2017-06-29 タタ コンサルタンシー サービシズ リミテッドＴＡＴＡ Ｃｏｎｓｕｌｔａｎｃｙ Ｓｅｒｖｉｃｅｓ Ｌｉｍｉｔｅｄ マルチセンサビジュアルアナリティクス

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06308008A (ja) * 1993-04-26 1994-11-04 Nippon Telegr & Teleph Corp <Ntt> 混合ガス成分判別および定量方法ならびに装置
JPH11264809A (ja) * 1998-03-17 1999-09-28 Shimadzu Corp ガス識別装置
JP2001305034A (ja) * 2000-04-26 2001-10-31 Mitsubishi Electric Corp ガス同定方法、ガス定量方法およびガス同定または定量装置
JP2011203006A (ja) * 2010-03-24 2011-10-13 Olympus Corp 検出センサ
US20160161433A1 (en) * 2014-05-16 2016-06-09 Massachusetts Institute Of Technology Electrospun Polymer Fibers for Gas Sensing
US20160084808A1 (en) * 2014-09-20 2016-03-24 Commissariat A L'energie Atomique Et Aux Energies Alternatives Method and device for determining a composition of a gas sample processed by means of gas chromatography
JP2017117438A (ja) * 2015-11-13 2017-06-29 タタ コンサルタンシー サービシズ リミテッドＴＡＴＡ Ｃｏｎｓｕｌｔａｎｃｙ Ｓｅｒｖｉｃｅｓ Ｌｉｍｉｔｅｄ マルチセンサビジュアルアナリティクス

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
SETO, SHUICHI ET AL.: "Chemical Analysis for transient response of semiconductor sensor by autoregressive model", IEEJ TRANSACTIONS ON SENSORS AND MICROMACHINES, vol. 125, no. 3, 1 June 2005 (2005-06-01), pages 129 - 134, XP055684041, DOI: 10.1541/ieejsmas.125.129 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPWO2021186532A1 (fr) * 2020-03-17 2021-09-23
WO2021186532A1 (fr) * 2020-03-17 2021-09-23 日本電気株式会社 Dispositif d'élimination de bruit, procédé d'élimination de bruit, et support d'enregistrement
US20230110600A1 (en) * 2020-03-17 2023-04-13 Nec Corporation Noise removing apparatus, noise removing method, and recording medium
JP7327649B2 (ja) 2020-03-17 2023-08-16 日本電気株式会社 ノイズ除去装置、ノイズ除去方法、及び、プログラム
US11846620B2 (en) 2020-03-17 2023-12-19 Nec Corporation Noise removing apparatus, noise removing method, and recording medium
US11974803B2 (en) 2020-10-12 2024-05-07 Biosense Webster (Israel) Ltd. Basket catheter with balloon

Also Published As

Publication number Publication date
JPWO2020026328A1 (ja) 2021-08-02

Similar Documents

Publication Publication Date Title
Dilthey et al. Strain-level metagenomic assignment and compositional estimation for long reads with MetaMaps
Karosiene et al. NetMHCcons: a consensus method for the major histocompatibility complex class I predictions
Bengtsson‐Palme et al. METAXA2: improved identification and taxonomic classification of small and large subunit rRNA in metagenomic data
Quinn et al. Understanding sequencing data as compositions: an outlook and review
Dong et al. A new taxonomy-based protein fold recognition approach based on autocross-covariance transformation
WO2017216980A1 (fr) Dispositif d&#39;apprentissage machine
JP7063389B2 (ja) 処理装置、処理方法、およびプログラム
JP6029192B2 (ja) 入力支援システム、入力支援方法および入力支援プログラム
WO2020026328A1 (fr) Dispositif de traitement d&#39;informations, procédé de commande et programme
CN110737689A (zh) 数据标准符合性检测方法、装置、系统及存储介质
Rapin et al. The MHC motif viewer: a visualization tool for MHC binding motifs
WO2020100285A1 (fr) Dispositif de traitement d&#39;informations, procédé de commande, et programme
WO2020065890A1 (fr) Dispositif de traitement d&#39;informations, procédé de traitement d&#39;informations et programme
MacNamara et al. T-cell epitope prediction: rescaling can mask biological variation between MHC molecules
JP7140191B2 (ja) 情報処理装置、制御方法、及びプログラム
CN111695743A (zh) 工厂化食品安全生产监控方法、装置、设备及存储介质
JP7099623B2 (ja) 情報処理装置、情報処理方法、およびプログラム
Rubin et al. Using ConSurf to detect functionally important regions in RNA
JP7056747B2 (ja) 情報処理装置、処理装置、情報処理方法、処理方法、決定方法、およびプログラム
JP7074194B2 (ja) 情報処理装置、制御方法、及びプログラム
Singh et al. SPOT-Contact-Single: Improving Single-Sequence-Based Prediction of Protein Contact Map using a Transformer Language Model
Raj et al. Significance of Sequence Features in Classification of Protein–Protein Interactions Using Machine Learning
CN107664655B (zh) 用于表征分析物的方法以及装置
Burks et al. POSMM: an efficient alignment-free metagenomic profiler that complements alignment-based profiling
JP3985826B2 (ja) 画像検索方法及びその装置

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: 18928154

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2020533926

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18928154

Country of ref document: EP

Kind code of ref document: A1