Information processing device, inference device, machine learning device, information processing method, inference method, and machine learning method

Abstract

This invention provides an information processing device that makes it possible to easily predict the likelihood of food-related accidents caused by marine products. [Solution] The system includes an information acquisition unit 400 that acquires aquaculture condition information, including aquaculture environment information indicating the environment of the aquaculture water area where the target aquatic product is cultivated, and a generation processing unit 401 that inputs the aquaculture condition information acquired by the information acquisition unit 400 into a learning model 12A that has been trained by machine learning to determine the correlation between aquaculture condition information for the target aquatic product and aquatic product hazard information indicating the presence of one or more types of hazardous substances contained in the aquatic product, thereby generating aquatic product hazard information for the target aquatic product.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an inference apparatus, a machine learning apparatus, an information processing method, an inference method, and a machine learning method.

Background Art

[0002] Conventionally, various devices and instruments have been developed for culturing aquatic products such as shellfish, seaweeds, fish, and crustaceans in waters such as the sea, rivers, and lakes (see, for example, Patent Document 1 and Patent Document 2).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] When aquatic products are eaten raw, if the aquatic products contain or are attached with harmful substances such as bacteria and viruses, there may occur a food accident (food poisoning) that has an adverse effect on the human body. However, since aquatic products are cultured in various waters and environments, it has been difficult to grasp the degree to which each aquatic product contains or is attached with harmful substances when the cultured aquatic products are caught.

[0005] In view of the above problems, an object of the present invention is to provide an information processing apparatus, an inference apparatus, a machine learning apparatus, an information processing method, an inference method, and a machine learning method that can easily predict the possibility of a food accident caused by aquatic products.

Means for Solving the Problems

[0006] To achieve the above objective, an information processing apparatus according to one aspect of the present invention is: An information acquisition unit that acquires aquaculture condition information, including aquaculture environment information that shows the environment of the aquaculture water area where the target fishery product is cultivated, The system includes a generation processing unit that generates the fishery product hazard information for the fishery product to be predicted by inputting the aquaculture condition information acquired by the information acquisition unit into a learning model that has been trained by machine learning to determine the correlation between the aquaculture condition information for the fishery product to be learned and fishery product hazard information indicating the presence of one or more types of hazardous substances contained in the fishery product, thereby generating the fishery product hazard information for the fishery product to be predicted. [Effects of the Invention]

[0007] According to an information processing device in one aspect of the present invention, the degree of presence of hazardous substances is predicted as fishery product hazard information for the fishery product in question, based on the fishery product hazard information output by inputting aquaculture condition information for the fishery product to be predicted into a learning model. Therefore, the possibility of food safety incidents caused by fishery products can be easily predicted.

[0008] Other issues, configurations, and effects will be clarified in the embodiments for carrying out the invention described later. [Brief explanation of the drawing]

[0009] [Figure 1] This is an overall configuration diagram showing an example of a fishery product aquaculture management system 1A according to the first embodiment. [Figure 2] This is a data configuration diagram showing an example of the aquaculture management database 22A according to the first embodiment. [Figure 3] This is a hardware configuration diagram showing an example of the Computer 900. [Figure 4] This is a block diagram showing an example of a machine learning device 3A according to the first embodiment. [Figure 5] This figure shows an example of a learning model 12A and training data 13A according to the first embodiment. [Figure 6] This flowchart shows an example of a machine learning method using machine learning device 3A. [Figure 7] It is a block diagram showing an example of the information processing apparatus 4A according to the first embodiment. [Figure 8] It is a functional explanatory diagram showing an example of the information processing apparatus 4A according to the first embodiment. [Figure 9] It is a flowchart showing an example of an information processing method by the information processing apparatus 4A. [Figure 10] It is an overall configuration diagram showing an example of the aquatic product aquaculture management system 1B according to the second embodiment. [Figure 11] It is a data configuration diagram showing an example of the aquaculture management database 22B according to the second embodiment. [Figure 12] It is a block diagram showing an example of the machine learning apparatus 3B according to the second embodiment. [Figure 13] It is a diagram showing an example of the learning model 12B and the learning data 13B according to the second embodiment. [Figure 14] It is a block diagram showing an example of the information processing apparatus 4B according to the second embodiment. [Figure 15] It is a functional explanatory diagram showing an example of the information processing apparatus 4B according to the second embodiment. [Figure 16] It is an overall configuration diagram showing an example of the aquatic product aquaculture management system 1C according to the third embodiment. [Figure 17A] It is a data configuration diagram showing an example of the aquaculture management database 22C according to the third embodiment. [Figure 17B] It is a data configuration diagram showing an example of the aquaculture management database 22C according to the third embodiment. [Figure 18] It is a block diagram showing an example of the machine learning apparatus 3C according to the third embodiment. [Figure 19] It is a diagram showing an example of the learning model 12C and the learning data 13C according to the third embodiment. [Figure 20] It is a block diagram showing an example of the information processing apparatus 4C according to the third embodiment. [Figure 21] It is a functional explanatory diagram showing an example of the information processing apparatus 4C according to the third embodiment.

Embodiments for Carrying Out the Invention

[0010] Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. Hereinafter, the scope necessary for the description to achieve the object of the present invention will be schematically shown, and mainly the scope necessary for the description of the relevant part of the present invention will be described, and the parts where the description is omitted shall be based on known techniques.

[0011] (First Embodiment) FIG. 1 is an overall configuration diagram showing an example of an aquatic product aquaculture management system 1A according to the first embodiment. The aquatic product aquaculture management system 1A according to the present embodiment manages the aquatic products 11 cultured in the aquaculture water area 10 and functions as a system for predicting the possibility of a food accident caused by the aquatic products 11 based on the aquaculture conditions when the aquatic products 11 are cultured in the aquaculture water area 10.

[0012] The aquaculture water area 10 is a water area for culturing various aquatic products 11. The aquaculture water area 10 may be, for example, either a seawater area or a freshwater area. The aquaculture water area 10 is not limited to a natural water area and may also be an artificial water area. The aquaculture water area 10 may be managed, for example, for each section delimited within a predetermined range as its management unit, or may be managed as a whole in a lump. <>

[0013] The aquatic product 11 is a shellfish, and examples include bivalves such as oysters, clams, scallops, and abalones. Examples of the spiral shellfish such as turban shells can be mentioned. In addition to the above-mentioned shellfish, the aquatic product 11 may also be seaweeds such as wakame, fish such as yellowtail, and crustaceans such as kuruma shrimp.

[0014] A hazardous substance is a substance that can cause food accidents that have adverse effects on the human body. Food accidents occur, for example, when raw seafood 11 containing or contaminated with a hazardous substance is consumed. Hazardous substances can be evaluated by detecting the presence of specific factors in the seafood 11, and these specific factors include, for example, bacteria and viruses. Specific examples of bacteria include total bacteria, coliform bacteria, Escherichia coli (E. coli), and Vibrio parahaemolyticus. Specific examples of viruses include norovirus. However, hazardous substances are not limited to the above examples as long as they can cause food accidents.

[0015] The aquaculture management system 1A predicts the likelihood of a food safety incident occurring when the target aquatic product 11 is consumed raw, based on information about the aquaculture conditions when the target aquatic product 11 is cultivated in the aquaculture area 10.

[0016] The aquaculture management system 1A, as shown in Figure 1, comprises, as its main components, an aquaculture management device 2A, a machine learning device 3A, an information processing device 4A, and a worker terminal device 5. Each device 2A to 5 is, for example, composed of a general-purpose or dedicated computer (see Figure 3 below) and connected to a wired or wireless network 6, enabling the mutual transmission and reception of various types of data. The number of each device 2A to 5 and the connection configuration of the network 6 are not limited to the example in Figure 1 and may be changed as appropriate.

[0017] The aquaculture management device 2A comprises an environmental sensor group 20 used to monitor the environment of the aquaculture area 10, a hazardous substance testing machine 21 used to test for hazardous substances contained in aquatic products 11 as samples, and an aquaculture management database 22A for managing aquaculture condition information indicating the aquaculture conditions for aquatic products 11 in the aquaculture area 10. The aquaculture management device 2A may also manage multiple aquaculture areas 10.

[0018] The environmental sensor group 20 consists of a meteorological environmental sensor that detects the weather environment of the aquaculture area 10, a water quality environmental sensor that detects the water quality environment of the aquaculture area 10, and a wave environmental sensor that detects the wave environment of the aquaculture area 10. The detection results from the environmental sensor group 20 are acquired as water area environmental information. The detection results from the meteorological environmental sensor are acquired as meteorological environmental information. The detection results from the water quality environmental sensor are acquired as water quality environmental information. The detection results from the wave environmental sensor are acquired as wave environmental information.

[0019] Weather environment sensors detect weather factors representing the weather environment, such as weather conditions, temperature, humidity, wind direction, wind speed, precipitation, and sunshine. Water quality environment sensors detect water quality factors representing the water quality environment, such as surface water temperature, underwater water temperature, dissolved oxygen level, chlorophyll a concentration, pH value, salinity, water level, and river flow rate. Wave environment sensors detect wave factors representing the wave environment, such as wave height, current direction, and current velocity.

[0020] Furthermore, the detection location (which may be multiple locations) and detection method of the environmental sensor group 20 can be appropriately determined according to the target to be detected. For example, image sensors (cameras) or weather satellites may be used. In addition, the environmental sensor group 20 may be installed not only in the aquaculture water area 10, but also in the surrounding environment at a predetermined distance from the aquaculture water area 10 (which may be multiple locations at different distances or in different directions). In this case, the environmental conditions may be those detected by the environmental sensor group 20 installed in the surrounding environment. Moreover, the aquatic environment information may be obtained from an information provision device managed by an external business operator, instead of or in addition to being obtained by the environmental sensor group 20.

[0021] The hazardous substance testing machine 21 is a device that consists of, for example, measuring instruments and testing kits, and is capable of testing the presence of one or more types of hazardous substances. If the hazardous substance testing machine 21 is configured to output the test results of hazardous substances as data, the aquaculture hazard information, including the test results of the hazardous substances, is registered in the aquaculture management database 22A. If the hazardous substance testing machine 21 does not have the function to output such aquaculture hazard information, the testing operator performs an input operation to input the test results of hazardous substances into the operator terminal device 5, and the aquaculture hazard information based on that input operation is registered in the aquaculture management database 22A.

[0022] The hazardous substance testing machine 21 may perform any of the following testing methods: morphological examination, biochemical characterization examination, serological examination, and genetic examination, or it may perform a combination of multiple testing methods. For example, the hazardous substance testing machine 21 may use tissue collected from fishery products 11 as a sample and perform flora analysis or genetic analysis of hazardous substances contained in the sample. For flora analysis, for example, next-generation sequencing (NGS) or real-time PCR (RT-PCR) can be used. In addition, although the type of hazardous substance to be tested for by the hazardous substance testing machine 21 is predetermined, multiple types of hazardous substance testing machines 21 with different test targets and testing methods may be used.

[0023] The aquaculture management database 22A registers the detection results from the environmental sensor group 20 (aquatic environment information), the inspection results of hazardous substances by the hazardous substance testing machine 21 (aquatic product hazard information), and so on. Further details of the aquaculture management database 22A will be described later.

[0024] The machine learning device 3A is a device that operates as the main component of the machine learning learning phase. For example, the machine learning device 3A acquires training data 13A from the aquaculture management database 22A and generates a learning model 12A used by the information processing device 4A based on the training data 13A. The trained learning model 12A is provided to the information processing device 4A via the network 6, recording media, etc.

[0025] The information processing device 4A is a device that operates as the main component of the machine learning inference phase. Using the learning model 12A generated by the machine learning device 3A, the information processing device 4A predicts the likelihood of a food safety incident occurring when the target aquatic product 11 is consumed raw, based on the aquaculture condition information of the target aquatic product 11. The predicted aquatic product hazard information is provided, for example, to the worker terminal device 5 and presented to inspection workers and aquaculture farm managers. Alternatively, the aquatic product hazard information may be provided to the aquaculture management device 2A and stored in the aquaculture management database 22A.

[0026] The worker terminal device 5 is a terminal device used by workers (aquaculture workers, inspection workers, aquaculture farm managers, test workers, etc.) working in the aquaculture area 10, and may be a stationary device or a portable device. The worker terminal device 5 accepts various input operations via a display screen such as an application program or a web browser or a reader / writer, and displays various information via the display screen.

[0027] Figure 2 is a data structure diagram showing an example of the aquaculture management database 22A according to the first embodiment. Note that each table also registers the time when each piece of information was acquired, but the time-related fields are omitted in Figure 2. Furthermore, Figure 2 shows examples of registration in each table's fields, including cases where the value is normalized to the range of 0 to 1, and cases where the class is registered, such as C1, C2, ..., Cn, etc.

[0028] The aquaculture management database 22A provides the aquaculture conditions for the aquaculture products 11 cultivated in the aquaculture water area 10. This is a database that manages fishery products 11 by type and management unit. The aquaculture management database 22A includes, for example, a fishery product table 220, a water environment table 221, and a fishery product hazard table 222, and is managed based on a fishery product ID that identifies the fishery product 11.

[0029] The aquatic product table 220 is used, for example, by aquaculture workers or aquaculture farm managers to input data using the worker terminal device 5, thereby registering the types of aquatic products 11 and the sections (management units) of the aquaculture water areas 10.

[0030] The aquatic environment table 221 registers aquatic environment information, including meteorological environment information, water quality environment information, and wave environment information, as detection results from the environmental sensor group 20. Note that aquatic environment information provided by external information providers may also be registered in the aquatic environment table 221.

[0031] The fishery product hazard table 222 registers the presence level of one or more types of hazardous substances (three types in the example in Figure 2) as a result of inspection by the hazardous substance inspection machine 21.

[0032] (Hardware configuration of each device) Figure 3 is a hardware configuration diagram showing an example of computer 900. Each of the devices 2A to 5 of the aquaculture management system 1A is configured with a general-purpose or dedicated computer 900.

[0033] As shown in Figure 3, the computer 900 comprises, as its main components, a bus 910, a processor 912, memory 914, an input device 916, an output device 917, a display device 918, a storage device 920, a communication interface unit 922, an external device interface unit 924, an I / O device interface unit 926, and a media input / output unit 928. Note that the above components may be omitted as appropriate depending on the intended use of the computer 900.

[0034] The processor 912 consists of one or more arithmetic processing units (CPU (Central Processing Unit), MPU (Micro-processing unit), DSP (digital signal processor), GPU (Graphics Processing Unit), etc.) and operates as a control unit that oversees the entire computer 900. The memory 914 stores various data and programs 930 and consists of volatile memory (DRAM, SRAM, etc.) that functions as main memory, and non-volatile memory (ROM), flash memory, etc.

[0035] The input device 916 consists of, for example, a keyboard, mouse, numeric keypad, electronic pen, microphone, etc., and functions as an input unit. The output device 917 consists of, for example, a sound (voice) output device, a vibration device, etc., and functions as an output unit. The display device 918 consists of, for example, a liquid crystal display, organic EL display, electronic paper, projector, etc., and functions as an output unit. The input device 916 and the display device 918 may be configured as an integrated unit, such as a touch panel display. The storage device 920 consists of, for example, an HDD, SSD, etc., and functions as a storage unit. The storage device 920 stores various data necessary for the execution of the operating system and program 930.

[0036] The communication I / F unit 922 is connected by wire or wireless to a network 940 such as the Internet or an intranet (which may be the same as network 6 in Figure 1) and functions as a communication unit that sends and receives data with other computers according to a predetermined communication standard. The external device I / F unit 924 is connected by wire or wireless to an external device 950 such as a camera, printer, scanner, or reader / writer and functions as a communication unit that sends and receives data with the external device 950 according to a predetermined communication standard. It functions as a communication unit that sends and receives data. The I / O device I / F unit 926 is connected to I / O devices 960 such as various sensors and actuators, and functions as a communication unit that sends and receives various signals and data to and from the I / O devices 960, such as detection signals from sensors and control signals to actuators. The media input / output unit 928 is composed of a drive device such as a DVD drive or CD drive, and reads and writes data to media (non-temporary storage medium) 970 such as DVDs and CDs.

[0037] In the computer 900 having the above configuration, the processor 912 calls and executes the program 930 stored in the storage device 920 in the memory 914, and controls various parts of the computer 900 via the bus 910. The program 930 may also be stored in the memory 914 instead of the storage device 920. The program 930 may be recorded on the media 970 in an installable or executable file format and provided to the computer 900 via the media input / output unit 928. The program 930 may also be provided to the computer 900 by downloading it via the network 940 through the communication interface unit 922. Furthermore, the computer 900 may implement the various functions realized by the processor 912 executing the program 930 using hardware such as an FPGA or ASIC.

[0038] Computer 900 is an electronic device of any form, consisting of, for example, a stationary computer or a portable computer. Computer 900 may be a client computer, a server computer, a cloud computer, or an embedded computer such as a control panel or controller (including microcontrollers, programmable logic controllers, and sequencers).

[0039] (Machine learning device 3A) Figure 4 is a block diagram showing an example of a machine learning device 3A according to the first embodiment. The machine learning device 3A includes a control unit 30, a communication unit 31, a training data storage unit 32, and a trained model storage unit 33.

[0040] The control unit 30 functions as a training data acquisition unit 300 and a machine learning unit 301. The communication unit 31 is connected to an external device via the network 6 and functions as a communication interface for sending and receiving various types of data.

[0041] The learning data acquisition unit 300 is connected to external devices (e.g., aquaculture management device 2A, worker terminal device 5, etc.) via the communication unit 31 and network 6, and acquires learning data 13A, which consists of aquaculture condition information as input data and fishery product hazard information as output data. The learning data 13A is used as training data, validation data, and test data in supervised learning. In addition, the fishery product hazard information is used as correct labels in supervised learning.

[0042] The learning data storage unit 32 is a database that stores multiple sets of learning data 13A acquired by the learning data acquisition unit 300. The specific configuration of the database constituting the learning data storage unit 32 can be designed as appropriate.

[0043] The machine learning unit 301 performs machine learning using multiple sets of training data 13A stored in the training data storage unit 32. Specifically, the machine learning unit 301 inputs multiple sets of training data 13A into the training model 12A and generates a trained training model 12A by having the training model 12A learn the correlation between the input data (aquaculture condition information) and output data (aquaculture product harm information) that constitute the training data 13A. Note that even if the machine learning unit 301 performs predetermined preprocessing on the input data (aquaculture condition information) input to the training model 12A, Alternatively, the output data (fishery product hazard information) output from the learning model 12A may be subjected to predetermined post-processing.

[0044] The trained model storage unit 33 is a database that stores the trained model 12A (specifically, the adjusted weight parameter set) generated by the machine learning unit 301. The trained model 12A stored in the trained model storage unit 33 is provided to the actual system (for example, the information processing device 4A) via the network 6 or a recording medium. In Figure 4, the training data storage unit 32 and the trained model storage unit 33 are shown as separate storage units, but they may be composed of a single storage unit.

[0045] Figure 5 shows an example of a learning model 12A and training data 13A according to the first embodiment. The training data 13A used for machine learning of the learning model 12A consists of input data and output data.

[0046] The input data constituting the training data 13A is aquaculture condition information that includes aquaculture environmental information indicating the environment of the aquaculture area 10 in which the fishery product 11 to be trained is cultivated. The aquaculture environmental information includes at least one of the following: meteorological environmental information, water quality environmental information, and wave environmental information of the aquaculture area 10. Preferably, the aquaculture environmental information includes at least meteorological environmental information, second preferably at least water quality environmental information, second preferably at least wave environmental information, and third preferably all of the above.

[0047] The meteorological environment information includes at least one meteorological factor from among weather, temperature, humidity, wind direction, wind speed, precipitation, and sunshine. The water quality environment information includes at least one water quality factor from among surface water temperature, underwater water temperature, dissolved oxygen level, chlorophyll a concentration, and pH value, salinity, water level, and river flow rate. The river flow rate is the flow rate of the river flowing into the aquaculture area 10. The wave environment information includes at least one wave factor from among wave height, current direction, and current velocity. The underwater water temperature is the water temperature at a predetermined depth, and may be multiple water temperatures at multiple depths.

[0048] Furthermore, the meteorological factors included in the meteorological environment information, the water quality factors included in the water quality environment information, and the wave factors included in the wave environment information may be time-series data showing the state at a predetermined point in time, or time-series data consisting of multiple time-series data at predetermined time intervals (every hour, every day, every week, every month, etc.). In addition, the meteorological factors included in the meteorological environment information, the water quality factors included in the water quality environment information, and the wave factors included in the wave environment information may be statistical values ​​such as the average, maximum, and minimum values ​​over a predetermined period, or statistical values ​​such as the average, maximum, and minimum values ​​at multiple detection points within the aquaculture area 10.

[0049] The output data constituting the training data 13A is fishery product hazard information indicating the presence of one or more types of hazardous substances contained in the fishery product 11 to be trained. The fishery product hazard information is at least one of the following: the presence of general bacteria, the presence of Escherichia coli, the presence of coliform bacteria, the presence of Vibrio parahaemolyticus, and the presence of norovirus.

[0050] The abundance of a hazardous substance can be expressed, for example, by its presence or absence, level of abundance, or number of individuals. The level of abundance can be a step value or a continuous value, and in the case of a continuous value, it may be a value normalized to a predetermined range (e.g., 0 to 1).

[0051] The learning data acquisition unit 300 acquires learning data 13A by referring to various information registered in the aquaculture management database 22A when the aquaculture products 11 were cultivated in the past, or by receiving input operations from the worker terminal device 5.

[0052] When the learning data acquisition unit 300 refers to the aquaculture management database 22A, for example, Training data 13A is obtained by acquiring aquaculture condition information and aquaculture product hazard information associated with a specific aquaculture product ID in the aquatic environment table 221 and the aquaculture product hazard table 222. If there are multiple aquaculture areas 10, the training data 13A may also be information acquired when aquaculture products 11 are cultivated under various aquaculture conditions in each aquaculture area 10. Alternatively, the training data 13A may also be information acquired when aquaculture products 11 are cultivated under various aquaculture conditions in a test area that mimics the aquaculture area 10.

[0053] The learning model 12A employs, for example, a neural network structure and comprises an input layer 120, a hidden layer 121, and an output layer 122. Synapses (not shown) connect each neuron between each layer, and each synapse is associated with a weight. The weight parameters, consisting of the weights of each synapse, are adjusted by machine learning.

[0054] The input layer 120 has a number of neurons corresponding to the aquaculture condition information as input data, and each value of the aquaculture condition information is input to each neuron. The output layer 122 has a number of neurons corresponding to the fishery product hazard information as output data, and the prediction result (inference result) of the fishery product hazard information for the aquaculture condition information is output as output data. If the learning model 12A is composed of a regression model, the fishery product hazard information is output as numerical values ​​normalized to a predetermined range (e.g., 0 to 1). If the learning model 12A is composed of a classification model, the fishery product hazard information is output as a score (accuracy) for each class, as numerical values ​​normalized to a predetermined range (e.g., 0 to 1).

[0055] In this embodiment, the data configuration of the learning model 12A and the learning data 13A was described as being configured as shown in Figure 5. However, multiple data configurations with different conditions may be adopted, for example, the machine learning method, the type of fishery product 11, the type of hazardous substance testing machine 21 (target of testing and testing method), the type of data included in the aquaculture condition information, the type of data included in the fishery product hazard information, etc. In that case, the learning data acquisition unit 300 acquires multiple types of learning data 13A corresponding to the multiple data configurations with different conditions, and the machine learning unit 301 performs machine learning using each of these learning data 13A.

[0056] (Machine learning methods) Figure 6 is a flowchart showing an example of a machine learning method using machine learning device 3A.

[0057] First, in step S100, the training data acquisition unit 300 acquires a desired number of training data 13A as preparation for starting machine learning, and stores the acquired training data 13A in the training data storage unit 32. The number of training data 13A prepared here should be set considering the inference accuracy required for the final learning model 12A.

[0058] Next, in step S110, the machine learning unit 301 prepares a pre-training model 12A in order to start machine learning. The pre-training model 12A prepared here consists of the neural network model exemplified in Figure 5, and the weights of each synapse are set to their initial values.

[0059] Next, in step S120, the machine learning unit 301 randomly selects, for example, one set of training data 13A from the multiple sets of training data 13A stored in the training data storage unit 32.

[0060] Next, in step S130, the machine learning unit 301 processes one set of training data 13A. The included aquaculture condition information (input data) is input to the input layer 120 of the prepared pre-training (or training) learning model 12A. As a result, the output layer 122 of the learning model 12A outputs fishery product hazard information (output data) as an inference result, but this output data is generated by the pre-training (or training) learning model 12A. Therefore, in the pre-training (or training) state, the output data output as an inference result shows information different from the fishery product hazard information (ground truth labels) included in the training data 13A.

[0061] Next, in step S140, the machine learning unit 301 compares the fishery product hazard information (ground truth labels) contained in the set of training data 13A acquired in step S120 with the fishery product hazard information (output data) output from the output layer 122 as an inference result in step S130, and performs machine learning by adjusting the weight of each synapse (backpropagation). In this way, the machine learning unit 301 trains the learning model 12A to recognize the correlation between aquaculture condition information and fishery product hazard information.

[0062] Next, in step S150, the machine learning unit 301 determines whether predetermined learning termination conditions have been met, for example, based on the evaluation value of the error function, which is based on the fishery product hazard information (correct labels) contained in the training data 13A and the fishery product hazard information output as an inference result (output data), or based on the remaining number of untrained training data 13A stored in the training data storage unit 32.

[0063] In step S150, if the machine learning unit 301 determines that the learning termination condition has not been met and that machine learning should continue (No in step S150), it returns to step S120 and repeats steps S120 to S140 multiple times using the untrained training data 13A on the learning model 12A that is currently being trained. On the other hand, in step S150, if the machine learning unit 301 determines that the learning termination condition has been met and that machine learning should be terminated (Yes in step S150), it proceeds to step S160.

[0064] Then, in step S160, the machine learning unit 301 stores the trained model 12A (set weight parameter set) generated by adjusting the weights associated with each synapse in the trained model storage unit 33, and the series of machine learning methods shown in Figure 6 is completed. In the machine learning method, step S100 corresponds to the training data storage step, steps S110 to S150 are the machine learning steps, and step S160 is the trained model storage step.

[0065] As described above, the machine learning device 3A and machine learning method according to this embodiment provide a learning model 12A that can predict (infer) information on fishery product hazards to the fishery product 11 from information on the cultivation conditions of the fishery product 11.

[0066] (Information Processing Device 4A) Figure 7 is a block diagram showing an example of an information processing device 4A according to the first embodiment. Figure 8 is a functional diagram showing an example of an information processing device 4A according to the first embodiment. The information processing device 4A comprises a control unit 40, a communication unit 41, and a learned model storage unit 42.

[0067] The control unit 40 functions as an information acquisition unit 400, a generation processing unit 401, and an output processing unit 402. The communication unit 41 is connected to an external device via the network 6 and functions as a communication interface for sending and receiving various types of data.

[0068] The information acquisition unit 400 is connected to external devices (e.g., aquaculture management device 2A, worker terminal device 5, etc.) via the communication unit 41 and network 6, and acquires aquaculture condition information, including aquaculture environment information that indicates the environment of the aquaculture water area 10 where the fishery products 11 to be predicted are cultivated.

[0069] For example, the information acquisition unit 400 acquires aquaculture condition information by receiving aquaculture environmental information from the environmental sensor group 20 as detection results from the environmental sensor group 20, which is the aquaculture environment information when the target aquatic product 11 is cultivated. Also, if aquaculture condition information for the target aquatic product 11 is registered in the aquaculture management database 22A, the information acquisition unit 400 acquires the aquaculture condition information by referring to the aquaculture management database 22A. In this case, the information acquisition unit 400 should acquire the aquaculture condition information in accordance with the data structure of the input data in the learning model 12A. Note that when the information acquisition unit 400 acquires time-series data of aquaculture environmental information, it may acquire current aquaculture environmental information from the environmental sensor group 20 and past aquaculture environmental information from the aquaculture management database 22A.

[0070] As described above, the generation processing unit 401 generates fishery product hazard information for the fishery product 11 based on the fishery product hazard information output by inputting the aquaculture condition information for the fishery product 11 acquired by the information acquisition unit 400 as input data to the learning model 12A. The generation processing unit 401 may perform predetermined pre-processing on the input data (aquaculture condition information) input to the learning model 12A, or it may perform predetermined post-processing on the output data (fishery product hazard information) output from the learning model 12A.

[0071] The trained model storage unit 42 is a database that stores trained models 12A used in the generation processing unit 401. The number of trained models 12A stored in the trained model storage unit 42 is not limited to the example above. For example, multiple trained models with different conditions may be stored and used selectively or in parallel, such as machine learning methods, types of fishery products 11, types of hazardous substance testing machines 21 (test targets and testing methods), types of data included in aquaculture condition information, and types of data included in fishery product hazard information. The trained model storage unit 42 may also be replaced by a storage unit on an external computer (for example, a server-type computer or a cloud-type computer). In that case, the generation processing unit 401 only needs to access the external computer.

[0072] The output processing unit 402 performs output processing to output the fishery product hazard information generated by the generation processing unit 401. For example, the output processing unit 402 may transmit the fishery product hazard information to the worker terminal device 5 so that a display screen based on the fishery product hazard information is displayed on the worker terminal device 5, or it may transmit the fishery product hazard information to the aquaculture management device 2A so that the fishery product hazard information is registered in the aquaculture management database 22A.

[0073] (Information processing methods) Figure 9 is a flowchart showing an example of an information processing method by the information processing device 4A. Below, we will describe an example of the operation of the information processing device 4A when an inspector checks the possibility of a food safety incident occurring without using the hazardous substance testing machine 21, using the fishery products 11 at the catch stage as the fishery products 11 to be predicted.

[0074] First, in step S200, the information acquisition unit 400 receives a fishery product ID that identifies the fishery product 11 to be predicted from the worker terminal device 5 used by the inspection worker. Then, the information acquisition unit 400 receives detection results from the environmental sensor group 20 that monitors the fishery product 11 to be predicted identified by the fishery product ID, or refers to the aquaculture management database 22A, thereby acquiring aquaculture condition information, including aquatic environment information, for the fishery product 11 to be predicted.

[0075] Next, in step S210, the generation processing unit 401 inputs the aquaculture condition information acquired in step S200 into the learning model 12A to generate aquaculture product hazard information for said aquaculture condition information as output data, and predicts the presence of hazardous substances contained in the target aquaculture product 11.

[0076] Next, in step S220, the output processing unit 402 transmits the fishery product hazard information generated in step S210 to the worker terminal device 5 used by the inspection worker as an output process for outputting the fishery product hazard information. The worker terminal device 5 then displays a screen based on the fishery product hazard information, presenting the fishery product hazard information for the fishery product 11 to the inspection worker. In addition to the worker terminal device 5, the fishery product hazard information may also be transmitted to the aquaculture management device 2A. In the above information processing method, step S200 corresponds to the information acquisition step, step S210 to the generation process step, and step S220 to the output process step.

[0077] As described above, according to the information processing device 4A and information processing method of this embodiment, by inputting aquaculture condition information for the fishery product 11 to be predicted into the learning model 12A, the degree of presence of hazardous substances contained in the fishery product 11 to be predicted is predicted as fishery product hazard information for the fishery product 11 to be predicted.

[0078] The growth status of aquatic products 11 is influenced by the meteorological, water quality, and wave conditions of the aquaculture area 10. For example, if the meteorological, water quality, and wave conditions are suitable for the growth of aquatic products 11, the growth status of the aquatic products 11 will be good, and the presence of hazardous substances will decrease. Therefore, the presence of hazardous substances in aquatic products 11 will fluctuate in accordance with the fluctuations in meteorological, water quality, and wave conditions information included in the aquaculture conditions information. Accordingly, by using these parameters that affect the growth status of aquatic products 11 as input data and inferring the presence of hazardous substances as output data, the presence of hazardous substances can be predicted with high accuracy.

[0079] (Second embodiment) Figure 10 is an overall configuration diagram showing an example of a fishery product aquaculture management system 1B according to the second embodiment. The second embodiment differs from the first embodiment in that the aquaculture management device 2B includes a pollution indicator substance testing machine 23 used to test for pollution indicator substances contained in water sampled from the aquaculture area 10. The second embodiment also differs from the first embodiment in that the aquaculture condition information, which constitutes the learning data 13B as input data, includes water pollution information indicating the degree of presence of pollution indicator substances contained in the aquaculture area 10. Below, the machine learning device 3B and information processing device 4B according to the second embodiment will be described, focusing on the differences from the first embodiment.

[0080] A contamination indicator substance is a substance that can serve as an indicator for evaluating the contamination level of the aquaculture water area 10. Contamination indicator substances can be evaluated by detecting the presence of specific factors in the aquaculture water area 10, and these specific factors include, for example, bacteria and viruses. Specific examples of bacteria include total bacteria, coliform bacteria, Escherichia coli (E. coli), and Vibrio parahaemolyticus. Specific examples of viruses include norovirus. However, contamination indicator substances are not limited to the above examples as long as they serve as an indicator of the contamination level of the aquaculture water area 10. Furthermore, a contamination indicator substance may be the same substance as the hazardous substance or a different substance.

[0081] The pollutant indicator substance testing machine 23, like the hazardous substance testing machine 21, is a device that can test the presence of one or more types of pollutant indicator substances, for example, by being composed of measuring instruments and test kits. If the pollutant indicator substance testing machine 23 is configured to output the test results of the pollutant indicator substances as data, the water pollution information including the test results of the pollutant indicator substances is registered in the aquaculture management database 22B. If the pollutant indicator substance testing machine 23 does not have the function to output the water pollution information as described above, the testing operator performs an input operation to input the test results of the pollutant indicator substances into the operator terminal device 5, and the water pollution information based on that input operation is registered in the aquaculture management database 22B.

[0082] The contamination indicator substance testing machine 23 may perform any of the following testing methods: morphological testing, biochemical characterization testing, serological testing, and genetic testing, or it may perform a combination of multiple testing methods. For example, the contamination indicator substance testing machine 23 may use water collected from the aquaculture area 10 as a sample and perform flora analysis or genetic analysis of the contamination indicator substances contained in the sample. Furthermore, although the types of contamination indicator substances to be tested for by the contamination indicator substance testing machine 23 are predetermined, multiple types of contamination indicator substance testing machines 23 with different targets and testing methods may be used. The water sample is collected, for example, at a location in the aquaculture area 10 that is geographically easy to collect from. In addition, the water sample may be collected not only from the aquaculture area 10, but also from related water areas associated with the aquaculture area 10, for example, from ocean currents or rivers that flow into the aquaculture area 10, or at locations that have an impact on the aquaculture area 10.

[0083] Figure 11 is a data configuration diagram showing an example of the aquaculture management database 22B according to the second embodiment. Similar to the first embodiment, the aquaculture management database 22B includes a fishery product table 220, a water environment table 221, and a fishery product hazard table 222, and further includes a water pollution table 223.

[0084] The water body pollution table 223 registers the presence levels of one or more types (three types in the example in Figure 11) of pollution indicator substances as a result of inspection by the pollution indicator substance inspector 23.

[0085] (Machine Learning Device 3B) Figure 12 is a block diagram showing an example of a machine learning device 3B according to the second embodiment. Figure 13 is a diagram showing an example of a learning model 12B and training data 13B according to the second embodiment.

[0086] The input data constituting the training data 13B, similar to the training data 13A according to the first embodiment, includes aquaculture condition information that includes aquaculture environmental information and aquaculture pollution information that indicates the presence of one or more types of pollution indicator substances contained in the aquaculture water area 10. The aquaculture pollution information includes at least one aquaculture pollution factor from among the presence of general bacteria, Escherichia coli, coliform bacteria, Vibrio parahaemolyticus, and norovirus contained in the aquaculture water area 10 or related water areas.

[0087] The abundance of an indicator of contamination can be expressed, for example, by the presence or absence of the indicator, its level of abundance, or the number of individuals present. The level of abundance can be a step value or a continuous value, and in the case of a continuous value, it may be a value normalized to a predetermined range (e.g., 0 to 1).

[0088] Furthermore, the water pollution information may include the results of flora analysis or genetic analysis of pollutant indicator substances contained in water samples taken from aquaculture waters 10 or related waters.

[0089] The output data constituting the training data 13B is fishery product hazard information, similar to the training data 13A in the first embodiment, so its explanation is omitted.

[0090] The learning data acquisition unit 300 acquires learning data 13B by referring to various information registered in the aquaculture management database 22B when the aquatic products 11 were cultivated in the past, or by receiving input operations from the worker terminal device 5. When the learning data acquisition unit 300 refers to the aquaculture management database 22B, for example, it acquires aquaculture condition information and aquatic product hazard information associated with a specific aquatic product ID in the water environment table 221, the aquatic product hazard table 222, and the water pollution table 223, thereby acquiring learning data Retrieve data 13B.

[0091] The machine learning unit 301 inputs multiple sets of training data 13B into the learning model 12B and generates a trained learning model 12B by having the learning model 12B learn the correlation between the input data (aquaculture condition information) and output data (aquaculture product damage information) that make up the training data 13B.

[0092] As described above, the machine learning device 3B and machine learning method according to this embodiment provide a learning model 12B that can predict (infer) information on fishery product hazards to the fishery product 11 from information on the cultivation conditions of the fishery product 11.

[0093] (Information Processing Device 4B) Figure 14 is a block diagram showing an example of the information processing device 4B according to the second embodiment. Figure 15 is a functional diagram showing an example of the information processing device 4B according to the second embodiment.

[0094] The information acquisition unit 400 receives aquaculture condition information from the environmental sensor group 20 as detection results from the environmental sensor group 20, when the fishery product 11 to be predicted is farmed, and also receives aquaculture condition information from the pollutant indicator substance tester 23 as inspection results from the pollutant indicator substance tester 23, when the fishery product 11 to be predicted is farmed, thereby acquiring aquaculture condition information. Furthermore, if aquaculture condition information and aquaculture condition information for the fishery product 11 to be predicted are registered in the aquaculture management database 22B, the information acquisition unit 400 acquires aquaculture condition information by referring to the aquaculture management database 22B. Note that the aquaculture condition information may also include the results of flora analysis or genetic analysis of pollutant indicator substances contained in water samples taken from the aquaculture water area 10 or related water areas.

[0095] As described above, the generation processing unit 401 generates fishery product hazard information for the fishery product 11 based on the fishery product hazard information output by inputting the aquaculture condition information for the fishery product 11 to be predicted, acquired by the information acquisition unit 400, as input data to the learning model 12B.

[0096] As described above, according to the information processing device 4B and information processing method of this embodiment, by inputting aquaculture condition information for the fishery product 11 to be predicted into the learning model 12B, the degree of presence of hazardous substances contained in the fishery product 11 to be predicted is predicted as fishery product hazard information for the fishery product 11.

[0097] The growth status of aquatic products 11 is influenced not only by the meteorological, water quality, and wave conditions of the aquaculture area 10, but also by the abundance of pollutant indicator substances contained in the aquaculture area 10 or related waters. For example, if the abundance of pollutant indicator substances in the aquaculture area 10 or related waters is high, the abundance of hazardous substances in the aquatic products 11 will also increase. Therefore, the abundance of hazardous substances in the aquatic products 11 fluctuates in accordance with the fluctuations in meteorological environment information, water quality environment information, wave environment information, and water pollution information included in the aquaculture conditions information. Accordingly, by using these parameters that affect the growth status of aquatic products 11 as input data and inferring the abundance of hazardous substances as output data, the abundance of hazardous substances can be predicted with high accuracy.

[0098] (Third embodiment) Figure 16 is an overall configuration diagram showing an example of a fishery product aquaculture management system 1C according to the third embodiment. The third embodiment differs in that the aquaculture management device 2C is connected to an information provision device 7 managed by an external business operator. Furthermore, in the third embodiment, the aquaculture condition information, which constitutes the learning data 13C, includes community infection status information showing the occurrence of food accidents in areas (cities) that discharge sewage flowing into the aquaculture water area 10, and the aquaculture water area 10 or This differs from the first embodiment in that it includes at least one piece of animal occurrence information indicating the occurrence of animals in related waters associated with the aquaculture area 10. Below, the machine learning device 3C and information processing device 4C according to the third embodiment will be described, focusing on the differences from the first embodiment.

[0099] Information provision device 7 is managed by businesses such as public and private organizations and provides community infection status information and animal habitat information to aquaculture management device 2C. Community infection status information and animal habitat information provided from information provision device 7 via network 6 are registered in aquaculture management database 22C.

[0100] Community transmission information includes infection levels for each type of hazardous substance, representing the occurrence of food-related incidents in the community. Infection levels are determined, for example, based on the number of infected individuals per unit population over a specified period. Infection levels can be stepped or continuous values; in the case of continuous values, they may be normalized to a specified range (e.g., 0 to 1).

[0101] Animal sighting information includes the sighting level for each animal species. Examples of animal species include mammals such as cats, and birds such as ducks, swans, and crows. The sighting level is determined, for example, based on the number of sightings in a given area over a given period. The sighting level can be a stepped value or a continuous value, and in the case of a continuous value, it may be a value normalized to a given range (e.g., 0 to 1).

[0102] Figures 17A and 17B are data configuration diagrams showing an example of the aquaculture management database 22C according to the third embodiment. Similar to the first embodiment, the aquaculture management database 22C has a fishery product table 220, a water environment table 221, and a fishery product hazard table 222, and further has a community infection status table 224 and an animal appearance status table 225.

[0103] The community transmission status table 224 registers the infection levels of one or more types (three types in the example of Figure 17B) of hazardous substances as community transmission status information provided by the information provision device 7. The animal appearance status table 225 registers the appearance levels of one or more types (three types in the example of Figure 17B) of animals as animal appearance status information provided by the information provision device 7.

[0104] (Machine learning device 3C) Figure 18 is a block diagram showing an example of a machine learning device 3C according to the third embodiment. Figure 19 is a diagram showing an example of a learning model 12C and training data 13C according to the third embodiment.

[0105] The input data constituting the training data 13C, similar to the training data 13A in the first embodiment, includes aquaculture condition information, as well as at least one of the following: community infection status information and animal appearance status information. The community infection status information includes infection levels for each type of hazardous substance. The animal appearance status information includes appearance levels for each type of animal.

[0106] The output data constituting the training data 13C is fishery product hazard information, similar to the training data 13A in the first embodiment, so its explanation is omitted.

[0107] The learning data acquisition unit 300 acquires learning data 13C by referring to various information registered in the aquaculture management database 22C when aquatic products 11 were farmed in the past, and by receiving input operations from the worker terminal device 5. When the learning data acquisition unit 300 refers to the aquaculture management database 22C, for example, the aquatic environment table 221, the aquatic product hazard table 222, the community infection status table 224, and the animal appearance status table. In cable 225, training data 13C is obtained by acquiring aquaculture condition information and fishery product hazard information associated with a specific fishery product ID.

[0108] The machine learning unit 301 inputs multiple sets of training data 13C into the learning model 12C and generates a trained learning model 12C by training the learning model 12C on the correlation between the input data (aquaculture condition information) and output data (aquaculture product damage information) that make up the training data 13C.

[0109] As described above, the machine learning device 3C and machine learning method according to this embodiment provide a learning model 12C that can predict (infer) information on fishery product hazards to the fishery product 11 from information on the cultivation conditions of the fishery product 11.

[0110] (Information Processing Device 4C) Figure 20 is a block diagram showing an example of an information processing device 4C according to the third embodiment. Figure 21 is a functional diagram showing an example of an information processing device 4C according to the third embodiment.

[0111] The information acquisition unit 400 receives aquatic environment information from the environmental sensor group 20 as detection results from the environmental sensor group 20, regarding the conditions under which the predicted aquatic product 11 is cultivated, and also receives community infection status information and animal appearance status information from the information providing device 7, thereby acquiring aquaculture condition information. Furthermore, if aquaculture condition information and aquatic pollution information for the predicted aquatic product 11 are registered in the aquaculture management database 22C, the information acquisition unit 400 acquires aquaculture condition information by referring to the aquaculture management database 22C.

[0112] As described above, the generation processing unit 401 generates fishery product hazard information for the fishery product 11 based on the fishery product hazard information output by inputting the aquaculture condition information for the fishery product 11 to be predicted, acquired by the information acquisition unit 400, as input data to the learning model 12C.

[0113] As described above, according to the information processing device 4C and information processing method of this embodiment, by inputting aquaculture condition information for the fishery product 11 to be predicted into the learning model 12C, the degree of presence of hazardous substances contained in the fishery product 11 to be predicted is predicted as fishery product hazard information for the fishery product 11.

[0114] For example, if sewage treatment is not carried out properly due to heavy rain or other reasons, hazardous substances contained in the sewage will flow into the aquaculture area 10, and the water quality of the aquaculture area 10 where the aquatic products 11 are cultivated will be affected by the infection level of the hazardous substances in the city. In addition, animals living in the aquaculture area 10 or related waters can become carriers of the hazardous substances, so the water quality of the aquaculture area 10 where the aquatic products 11 are cultivated will be affected by the presence level of animals living in the aquaculture area 10 or related waters. Therefore, the growth status of the aquatic products 11 and the water quality of the aquaculture area 10 are affected not only by the weather environment, water quality environment and wave environment of the aquaculture area 10, but also by the infection level of the hazardous substances in the city and the presence level of animals in the aquaculture area 10 or related waters. By using these as input data and inferring the presence of the hazardous substances as output data, the presence of the hazardous substances can be predicted with high accuracy.

[0115] (Other embodiments) The present invention is not limited to the embodiments described above, and can be implemented with various modifications without departing from the spirit of the invention. All such modifications are included in the technical concept of the present invention.

[0116] The above embodiments may be combined as appropriate. For example, by combining the second embodiment and the third embodiment, the environmental condition information may include at least one of the following: aquatic environment information, aquatic pollution information, community infection status information, and animal appearance status information. In that case, the data structure of the learning model and the learning data may be changed as appropriate.

[0117] In the above embodiment, the information processing devices 4A to 4C were described as predicting fishery product hazard information for the time when aquaculture condition information is acquired (present time). However, they may also predict fishery product hazard information for a future time (future time). In this case, the information processing devices 4A to 4C may predict both the fishery product hazard information for the present time and the fishery product hazard information for a future time, or they may predict the change in fishery product hazard information over time by predicting fishery product hazard information for multiple future time points (for example, every week or every month). In that case, the data configuration of the learning models 12A to 12C and the learning data 13A to 13C should be changed as appropriate.

[0118] In the above embodiment, the aquaculture management devices 2A-2C, the machine learning devices 3A-3C, and the information processing devices 4A-4C were described as being composed of separate devices. However, these three devices may be composed of a single device, or any two of these three devices may be composed of a single device. Furthermore, at least one of the machine learning devices 3A-3C and the information processing devices 4A-4C may be incorporated into the worker terminal device 5.

[0119] In the above embodiment, we have described the case in which neural networks are used as the learning models 12A to 12C that realize machine learning by the machine learning unit 301, but other machine learning models may also be used. Other machine learning models include, for example, tree-type models such as decision trees and regression trees, ensemble learning such as bagging and boosting, recurrent neural networks, convolutional neural networks, and neural network types such as LSTM (deep (including learning), hierarchical clustering, non-hierarchical clustering, k-nearest neighbors, k-squared Examples include clustering methods such as equality analysis, multivariate analysis such as principal component analysis, factor analysis, and logistic regression, and support vector machines.

[0120] In the above embodiment, the output processing unit 402 of the information processing devices 4A to 4C was described as performing output processing to output the fishery product hazard information generated by the generation processing unit 401. However, the fishery product hazard information generated by the generation processing unit 401 may be used for any purpose other than those described in the above embodiment. For example, the fishery product hazard information generated by the generation processing unit 401 may be used to determine the processing conditions before shipping the fishery product 11.

[0121] Specifically, the control unit 40 of the information processing devices 4A to 4C functions as a processing condition determination unit (processing condition determination process) that determines the processing conditions before shipment based on the marine product hazard information generated by the generation processing unit 401, similar to the respective units 400 to 402.

[0122] The processing condition determination unit determines, as a processing condition before shipment, whether to perform purification treatment in a tank, high-pressure treatment with high-pressure washing water, or a combination of purification treatment and high-pressure treatment, based on, for example, the abundance of each hazardous substance contained in the marine product 11. At that time, the processing condition determination unit determines, as a detailed condition for purification treatment, for example, the purification time according to the abundance of each hazardous substance, and as a detailed condition for high-pressure treatment, for example, the water pressure, water temperature, and washing time of the high-pressure washing water according to the abundance of each hazardous substance.

[0123] (Machine learning programs and information processing programs) The present invention can also be provided in the form of a program (machine learning program) that causes the computer 900 to function as each part of the machine learning device 3A to 3C, or a program (machine learning program) that causes the computer 900 to execute each step of the machine learning method. Furthermore, the present invention can also be provided in the form of a program (information processing program) for causing the computer 900 to function as each part of the information processing apparatus 4A to 4C, or a program (information processing program) for causing the computer 900 to execute each step of the information processing method according to the above embodiment.

[0124] (Inference device, inference method, and inference program) The present invention can be provided not only in the form of the information processing devices 4A to 4C (information processing methods or information processing programs) according to the above embodiments, but also in the form of an inference device (inference method or inference program) used to infer information on damage to aquatic products. In that case, the inference device (inference method or inference program) may include a memory and a processor, the processor of which may execute a series of processes. The series of processes includes an information acquisition process (information acquisition step) for acquiring aquaculture condition information for the aquatic product 11 to be predicted, and an inference process (inference step) for inferring information on damage to aquatic products 11 after acquiring the aquaculture condition information in the information acquisition process.

[0125] By providing the inference device (inference method or inference program) in the form of an inference device, it becomes easier to apply to various devices compared to implementing an information processing device. It will be obvious to those skilled in the art that when the inference device (inference method or inference program) infers information on fishery product hazards, it may apply the inference method performed by the generation processing unit using the trained learning model generated by the machine learning device and machine learning method according to the above embodiment.

[0126] The various aspects of this disclosure are summarized below as an appendix.

[0127] (Note 1) An information acquisition unit that acquires aquaculture condition information, including aquaculture environment information that shows the environment of the aquaculture water area where the target fishery product is cultivated, The system includes a generation processing unit that generates the fishery product hazard information for the fishery product to be predicted by inputting the aquaculture condition information acquired by the information acquisition unit into a learning model that has learned the correlation between the aquaculture condition information for the fishery product to be learned and fishery product hazard information indicating the presence of one or more types of hazardous substances contained in the fishery product, using machine learning. Information processing device. (Note 2) The aforementioned aquatic environment information is, Meteorological environmental information indicating the meteorological environment of the aquaculture area, Water quality environmental information indicating the water quality environment of the aquaculture area, and Includes at least one of the wave environment information indicating the wave environment of the aquaculture area, The information processing device described in Appendix 1. (Note 3) The aforementioned meteorological environmental information is, weather, temperature, humidity, wind direction, wind speed, Precipitation, and, Includes at least one meteorological factor among the amount of sunlight, The aforementioned water quality environmental information is, water surface temperature, Underwater temperature, Dissolved oxygen content, Chlorophyll a concentration, pH value, Salt concentration Water level, and, The river flow includes at least one water quality factor, The aforementioned wave environment information is, Wave height, Current direction, and, The flow velocity includes at least one wave factor, The information processing device described in Appendix 2. (Note 4) The aforementioned aquaculture condition information is This includes aquatic pollution information indicating the presence of one or more types of pollutant indicator substances in the aquaculture waters or related waters associated with the aquaculture waters, An information processing device as described in any one of the appendices 1 to 3. (Note 5) The aforementioned water body pollution information is, The abundance of general bacteria contained in the aforementioned aquaculture waters, The abundance of E. coli contained in the aforementioned aquaculture waters, The abundance of coliform bacteria in the aforementioned aquaculture waters, The presence of Vibrio parahaemolyticus in the aforementioned aquaculture waters, and The presence of norovirus in the aforementioned aquaculture waters includes at least one water pollution factor, The information processing device described in Appendix 4. (Note 6) The aforementioned water body pollution information is, Using water collected from the aquaculture area or related area as a sample, the analysis results of flora analysis or genetic analysis of the pollution indicator substances contained in the sample include: The information processing device described in Appendix 4 or Appendix 5. (Note 7) The aforementioned aquaculture condition information is This includes community transmission information showing the occurrence of food-related incidents in areas that discharge sewage into the aforementioned aquaculture waters, The information processing apparatus according to claim 1. (Note 8) The aforementioned aquaculture condition information is This includes animal sighting information showing the animal sighting status in the aquaculture area or related waters associated with the aquaculture area, An information processing device as described in any one of the appendices 1 to 7. (Note 9) The aforementioned information regarding harm to fishery products is, The abundance of general bacteria contained in the aforementioned fishery product, The amount of E. coli present in the aforementioned fishery product, The degree of presence of coliform bacteria in the aforementioned fishery products, The presence of Vibrio parahaemolyticus in the aforementioned fishery product, and At least one of the norovirus levels present in the aforementioned seafood is: An information processing device as described in any one of the appendices 1 to 8. (Note 10) The system includes a processing condition determination unit that determines the processing conditions for the fishery products before shipment based on the fishery product hazard information generated by the generation processing unit. An information processing device as described in any one of the appendices 1 to 9. (Note 11) The processing condition determination unit is, Based on the degree of the hazardous substance contained in the aquatic product, it is determined whether to perform purification treatment in a tank, high-pressure treatment with high-pressure washing water, or a combination of the purification treatment and the high-pressure treatment. The information processing device described in Appendix 10. [Explanation of Symbols]

[0128] 1A-1C: Aquaculture management system, 2A-2C: Aquaculture management equipment, 3A-3C: Machine learning devices, 4A-4C: Information processing devices 5...Worker terminal device, 6...Network, 10...Aquaculture area, 11...Fishery products, 12A-12C...Learning model, 13A-13C...Training data, 20…Environmental sensor group, 21…Hazardous substance detection machine, 22A~22C…Aquaculture management database, 23…Contamination indicator substance testing machine, 30...Control unit, 31...Communication unit, 32...Learning data storage unit, 33...Trained model memory unit, 40...Control unit, 41...Communication unit, 42...Learned model memory unit, 300...Training data acquisition unit, 301...Machine learning unit, 400... Information acquisition unit, 401... Generation processing unit, 402... Output processing unit, 900... Computer

Claims

1. An information acquisition unit that acquires aquaculture condition information, including aquaculture environment information that shows the environment of the aquaculture water area where the target fishery product is cultivated, The system includes a generation processing unit that generates the fishery product hazard information for the fishery product to be predicted by inputting the aquaculture condition information acquired by the information acquisition unit into a learning model that has learned the correlation between the aquaculture condition information for the fishery product to be learned and fishery product hazard information indicating the presence of one or more types of hazardous substances contained in the fishery product, by machine learning, and a learning model that generates the fishery product hazard information for the fishery product to be predicted. Information processing device.

2. The aforementioned aquatic environment information is, Meteorological environmental information indicating the meteorological environment of the aquaculture area, Water quality environmental information indicating the water quality environment of the aquaculture area, and This includes at least one wave environment information item indicating the wave environment of the aquaculture area, The information processing apparatus according to claim 1.

3. The aforementioned meteorological environmental information is, weather, temperature, humidity, wind direction, wind speed, Precipitation, and, It includes at least one meteorological factor among the amount of sunlight, The aforementioned water quality environmental information is, water surface temperature, Underwater temperature, Dissolved oxygen content, Chlorophyll a concentration, pH value, Salt concentration Water level, and, The river flow includes at least one water quality factor, The aforementioned wave environment information is, Wave height, Current direction, and, The flow velocity includes at least one wave factor, The information processing apparatus according to claim 2.

4. The aforementioned aquaculture condition information is This includes aquatic pollution information indicating the presence of one or more types of pollutant indicator substances in the aquaculture waters or related waters associated with the aquaculture waters, The information processing apparatus according to claim 1.

5. The aforementioned water body pollution information is, The abundance of general bacteria contained in the aforementioned aquaculture waters, The abundance of E. coli contained in the aforementioned aquaculture waters, The abundance of coliform bacteria in the aforementioned aquaculture waters, The presence of Vibrio parahaemolyticus in the aforementioned aquaculture waters, and The presence of norovirus in the aforementioned aquaculture waters is determined by at least one water pollution factor. including, The information processing apparatus according to claim 4.

6. The aforementioned water body pollution information is, Using water collected from the aquaculture area or related area as a sample, the analysis results of flora analysis or genetic analysis of the pollution indicator substances contained in the sample include: The information processing apparatus according to claim 4.

7. The aforementioned aquaculture condition information is This includes community transmission information showing the occurrence of food-related incidents in areas that discharge sewage into the aforementioned aquaculture waters, The information processing apparatus according to claim 1.

8. The aforementioned aquaculture condition information is This includes animal sighting information showing the animal sighting status in the aquaculture area or related waters associated with the aquaculture area, The information processing apparatus according to claim 1.

9. The aforementioned information regarding harm to fishery products is, The abundance of general bacteria contained in the aforementioned fishery product, The amount of E. coli present in the aforementioned fishery product, The degree of presence of coliform bacteria in the aforementioned fishery products, The presence of Vibrio parahaemolyticus in the aforementioned fishery product, and At least one of the norovirus levels present in the aforementioned seafood product, The information processing apparatus according to claim 1.

10. The system includes a processing condition determination unit that determines the processing conditions for the fishery products before shipment based on the fishery product hazard information generated by the generation processing unit. The information processing apparatus according to claim 1.

11. The processing condition determination unit is, Based on the degree of the hazardous substance contained in the aquatic product, it is determined whether to perform purification treatment in a tank, high-pressure treatment with high-pressure washing water, or a combination of the purification treatment and the high-pressure treatment. The information processing apparatus according to claim 10.

12. An inference device comprising memory and a processor, The aforementioned processor, Information acquisition process to acquire aquaculture condition information, including aquaculture environment information that shows the environment of the aquaculture water area where the target fishery product is cultivated, When the aquaculture condition information is obtained in the information acquisition process, an inference process is performed to infer aquatic product hazard information indicating the presence of one or more types of hazardous substances contained in the aquatic product to be predicted. Reasoning device.

13. A learning data storage unit that stores multiple sets of learning data consisting of input data and output data, A machine learning unit that inputs multiple sets of the aforementioned training data into the learning model to train the learning model on the correlation between the input data and the output data, The machine learning unit stores the learned model in which the correlation has been learned. Equipped with a Dell memory unit, The aforementioned input data is This is aquaculture condition information that includes aquatic environment information showing the environment of the aquaculture water area where the target fishery product is farmed. The output data mentioned above is: This is fishery product hazard information indicating the degree of presence of one or more hazardous substances contained in the fishery product being studied. Machine learning device.

14. A method of information processing performed by a computer, An information acquisition process to acquire aquaculture condition information, including aquaculture environmental information that shows the environment of the aquaculture water area where the target fishery product is cultivated, The system includes a generation process that generates the fishery product hazard information for the fishery product to be predicted by inputting the aquaculture condition information obtained in the information acquisition process into a learning model that has been trained by machine learning to recognize the correlation between the aquaculture condition information for the fishery product to be learned and fishery product hazard information indicating the presence of one or more types of hazardous substances contained in the fishery product, thereby generating the fishery product hazard information for the fishery product to be predicted. Information processing methods.

15. An inference method performed by an inference device comprising memory and a processor, The aforementioned processor, Information acquisition process to acquire aquaculture condition information, including aquaculture environment information that shows the environment of the aquaculture water area where the target fishery product is cultivated, When the aquaculture condition information is obtained in the information acquisition process, an inference process is performed to infer aquatic product hazard information indicating the presence of one or more types of hazardous substances contained in the aquatic product to be predicted. Reasoning method.

16. A machine learning method performed by a computer, A learning data storage process that stores multiple sets of learning data consisting of input data and output data, A machine learning process in which multiple sets of the aforementioned training data are input into the training model to train the training model to learn the correlation between the input data and the output data, The system includes a trained model storage step for storing the trained model that has learned the correlation relationship through the machine learning step, The aforementioned input data is This is aquaculture condition information that includes aquatic environment information showing the environment of the aquaculture water area where the target fishery product is farmed. The output data mentioned above is: This is fishery product hazard information indicating the degree of presence of one or more hazardous substances contained in the fishery product being studied. Machine learning methods.

Images