CN113113089A - Smell identification method based on big data analysis - Google Patents
Smell identification method based on big data analysis Download PDFInfo
- Publication number
- CN113113089A CN113113089A CN202110414388.XA CN202110414388A CN113113089A CN 113113089 A CN113113089 A CN 113113089A CN 202110414388 A CN202110414388 A CN 202110414388A CN 113113089 A CN113113089 A CN 113113089A
- Authority
- CN
- China
- Prior art keywords
- odor
- data
- model
- substance
- big data
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/20—Identification of molecular entities, parts thereof or of chemical compositions
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Investigating Or Analyzing Materials By The Use Of Electric Means (AREA)
Abstract
The invention provides a smell identification method based on big data analysis, which is characterized in that a corresponding standard smell molecule database is established by collecting chemical molecule structure information data in a big data platform, and data preprocessing is carried out by combining with a chemical theory to extract proper characteristics. According to different chemical structure characteristics of different odor substances, similarity classification is carried out on the different odor substances by adopting algorithms such as KNN (Konnen) and BP (Back propagation) neural networks and the like to establish a model, and the recognition rate of the odor is compared, so that the odor of the substance is better recognized through the chemical structure of the substance. When the classification recognition rate is higher, the extracted features can effectively distinguish the substance odor information, namely, the extracted features are closer to the essence of the substance odor and can better represent the odor information of the substance. The method is stable and simple, can provide different thought references for odor identification, can save a great deal of energy and investment required by searching specific odor molecules, and lays a foundation for further researching the odor universality characterization mode of the substance.
Description
Technical Field
The invention relates to a smell identification method based on big data analysis.
Background
The flavor is a complex and multi-sensory human experience and has a rich evolutionary history. The molecules form the chemical basis of flavor, and are expressed primarily by the mechanisms of taste and smell. The sense of smell is different from the sense of taste and is difficult to define in a simple manner. This also leads to another very painful problem for scientists-how to predict odor.
The traditional smell identification method only can directly measure human smell, so that a large amount of time and material resources are needed to cultivate experts specially used for identifying the smell. However, the manual identification has a large subjective factor, and in a sense, the judgment result of the sensory evaluation method has considerable individual difference depending on the identification person because of the influence of subjective factors such as experience, emotion and the like, and in addition, the human sense organ cannot be used for detecting toxic gas, continuous work and remote operation.
In recent years, artificial intelligence olfaction has been greatly developed worldwide.
The common machine olfaction system collects sensor data by using an electronic nose, and odor molecules are adsorbed by a sensor array in the machine olfaction system to generate electric signals; the generated signals are processed and transmitted by various methods, and the processed signals are judged and recognized by a computer mode recognition system, but due to the particularity of gas data, the signals are easily affected by the phenomena of temperature, humidity, sensor drift and the like in the data acquisition process, and finally, the experimental result may have deviation.
Disclosure of Invention
The invention aims to provide an odor identification method based on big data analysis.
In order to solve the above problems, the present invention provides a method for identifying odors based on big data analysis, comprising:
acquiring odor molecular structure characteristic information of different substances on a big data platform through a crawler technology, and selecting the odor molecular structure characteristic information of representative and multidimensional chemical molecules to establish a self-defined odor molecular database;
carrying out corresponding data preprocessing on the odor molecule structure characteristic information of the odor molecule database;
performing feature selection and feature extraction on the odor molecular structure feature information subjected to data preprocessing to extract odor comprehensive features which can most represent different substances;
performing feature selection and feature extraction on the odor molecular structure feature information subjected to data preprocessing to extract odor comprehensive features which can most represent different substances;
and establishing a model based on comprehensive odor characteristics which can best represent different substances so as to predict the odor of the substances through the model.
Further, in the above method, building a model based on the comprehensive characteristics of odors that can best characterize different substances to predict the odors of the substances through the model, the method includes:
taking the extracted comprehensive odor characteristics which can represent different substances most as the input of a model, and carrying out training test by continuously adjusting parameters to obtain an optimized model;
and (4) predicting the odor of the unknown sample on the optimized model.
Further, in the above method, the feature selection and feature extraction are performed on the odor molecule structure feature information after the data preprocessing, so as to extract the comprehensive odor features that can best represent different substances, including:
and performing dimensionality reduction treatment on the multi-dimensional odor molecular structure characteristic information by adopting a PCA algorithm in combination with actual chemical correlation knowledge, and extracting odor comprehensive characteristics which can represent different substances most.
Further, in the above method, the information on the structural characteristics of the odor molecule includes: molecular weight, atomic number, and electrochemical properties.
Further, in the above method, the data preprocessing includes: data cleaning and abnormal data elimination.
Further, in the above method, modeling to predict the odor of the substance includes:
and (3) establishing a model by adopting an algorithm of the KNN and BP neural networks to predict the odor of the substance.
Further, in the above method, an algorithm of the KNN and BP neural networks is used to build a model to predict the odor of the substance, including:
step S401: loading comprehensive odor characteristics which can represent different substances most as a data set, and dividing the data set into a training set and a testing set;
step S402: calculating a distance value between the test data in the test set and the training data in each training set by using the Manhattan distance;
step S403: sorting according to the increasing relation of the distance values;
step S404: selecting proper K values by a cross validation method, and finding out categories corresponding to the test data with the shortest K distance values;
step S405: and determining the occurrence frequency of the category where the first K points are located, and returning the category with the highest occurrence frequency in the first K test data as the prediction classification of the test data.
Compared with the prior art, the method has the advantages that the corresponding standard odor molecule database is established by collecting chemical molecule structure information data in the big data platform, and the data preprocessing is carried out by combining with the chemical theory to extract proper characteristics. According to different chemical structure characteristics of different odor substances, similarity classification is carried out on the different odor substances by adopting algorithms such as KNN (Konnen) and BP (Back propagation) neural networks and the like to establish a model, and the recognition rate of the odor is compared, so that the odor of the substance is better recognized through the chemical structure of the substance. When the classification recognition rate is higher, the extracted features can effectively distinguish the substance odor information, namely, the extracted features are closer to the essence of the substance odor and can better represent the odor information of the substance. The method is stable and simple, can provide different thought references for odor identification, can save a great deal of energy and investment required by searching specific odor molecules, and lays a foundation for further researching the odor universality characterization mode of the substance.
Drawings
Fig. 1 is a flowchart of a big data analysis-based scent recognition method according to an embodiment of the present invention;
fig. 2 is a flow chart of KNN algorithm provided in the embodiment of the present invention;
fig. 3 is a schematic diagram of a BP neural network algorithm provided in the embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in fig. 1, a method for identifying odors based on big data analysis includes the following steps:
step S1: acquiring odor molecular structure characteristic information of different substances on a big data platform through a crawler technology, and selecting the odor molecular structure characteristic information of representative and multidimensional chemical molecules to establish a self-defined odor molecular database;
the collected data is constructed into a self-defined smell molecule database, so that the storage and subsequent use of smell data are greatly facilitated;
acquiring enough chemical structure characteristic information of different odors from a big data platform, wherein the information comprises the atom number, molecular weight, electrochemical characteristics and the like of odor molecules, and storing the original data into a custom odor database so as to facilitate the storage and subsequent use of odor data;
step S2: carrying out corresponding data preprocessing on the odor molecule structure characteristic information of the odor molecule database;
here, a series of data preprocessing is required to be performed on the complex multidimensional data collected in the database, including data cleaning, abnormal data elimination and other operations, after the original odor data is preprocessed, the most effective feature vector needs to be found from the original features to represent the odor information of odor molecules, that is, feature selection and feature extraction are required to be performed later, which directly affects the accuracy and stability of a subsequently established model, therefore, a PCA algorithm is adopted to perform dimension reduction processing on the multidimensional data in combination with actual chemical theory knowledge, comprehensive features which can represent different odor types most are extracted, and a visualization result is obtained;
step S3: performing feature selection and feature extraction on the odor molecular structure feature information subjected to data preprocessing to extract odor comprehensive features which can most represent different substances;
step S4: and establishing a model based on comprehensive odor characteristics which can best represent different substances so as to predict the odor of the substances through the model.
Here, algorithms such as KNN and BP neural networks can be used to establish a model to predict the odor type of a substance, the data set processed in the step 2 is divided into a training set and a test set, the training set is used to train the model, the test set is used to verify the quality of the model, and the experiment is performed according to the following formula 8: 2 was tested. And (3) taking the feature vector extracted in the step (2) as the input of the model, and performing training test by continuously adjusting parameters to select the most appropriate odor prediction model.
Finally, outputting the type of the unknown odor molecules by inputting the characteristic vectors of the unknown odor molecules on an optimized model, and realizing the odor prediction. The method is stable and simple, can provide different thought references for odor identification, can save a great deal of energy and investment required by searching specific odor molecules, and lays a foundation for further researching the odor universality characterization mode of the substance.
Aiming at the defects in the prior art, the invention provides a smell identification method based on big data analysis, which predicts the smell according to the chemical structure of the molecule, solves the problem that the deviation of the current artificial olfaction direct measurement and the machine olfaction collection data are inaccurate, improves the identification rate in the smell identification and reduces the deviation.
In an embodiment of the smell recognition method based on big data analysis, step S4: establishing a model based on comprehensive odor characteristics which can best represent different substances so as to predict the odor of the substances through the model, wherein the model comprises the following steps:
taking the extracted comprehensive odor characteristics which can represent different substances most as the input of a model, and carrying out training test by continuously adjusting parameters to obtain an optimized model;
and (4) predicting the odor of the unknown sample on the optimized model.
In an embodiment of the smell recognition method based on big data analysis, step S3: the method for extracting the comprehensive odor characteristics most capable of representing different substances by performing characteristic selection and characteristic extraction on the odor molecular structure characteristic information subjected to data preprocessing comprises the following steps:
and performing dimensionality reduction treatment on the multi-dimensional odor molecular structure characteristic information by adopting a PCA algorithm in combination with actual chemical correlation knowledge, and extracting odor comprehensive characteristics which can represent different substances most.
Here, in step S3, a PCA algorithm is used for feature selection and feature extraction of the high-dimensional data.
In an embodiment of the big data analysis-based odor identification method of the present invention, the odor molecular structure feature information includes: molecular weight, atomic number, and electrochemical properties.
In an embodiment of the big data analysis-based odor identification method of the present invention, the data preprocessing includes: data cleaning and abnormal data elimination.
In an embodiment of the big data analysis-based odor identification method of the present invention, establishing a model to predict the odor of a substance includes:
and (3) establishing a model by adopting an algorithm of the KNN and BP neural networks to predict the odor of the substance.
Fig. 3 is a schematic diagram of a BP neural network algorithm provided in the embodiment of the present invention.
In an embodiment of the big data analysis-based odor identification method, the model is established by adopting an algorithm of KNN and BP neural networks to predict the odor of the substance, and the method comprises the following steps:
step S401: loading comprehensive odor characteristics which can represent different substances most as a data set, and dividing the data set into a training set and a testing set;
step S402: calculating a distance value between the test data in the test set and the training data in each training set by using the Manhattan distance;
step S403: sorting according to the increasing relation of the distance values;
step S404: selecting proper K values by a cross validation method, and finding out categories corresponding to the test data with the shortest K distance values;
step S405: and determining the occurrence frequency of the category where the first K points are located, and returning the category with the highest occurrence frequency in the first K test data as the prediction classification of the test data.
In an embodiment of the method for identifying odor based on big data analysis, the categories are divided into five categories, and for convenience of result prediction, the odor categories can be classified into sweet taste, sour taste, stink taste, fruit taste and flower taste.
In an embodiment of the big data analysis-based odor identification method of the present invention, a ratio of the training set to the test set is 8: 2.
before the model is established, a data set is divided into a training set and a testing set, the training set is used for training the model, the testing set is used for verifying the quality of the model, and the experimental period is as follows: 2 was tested.
Fig. 2 is a flowchart of the KNN algorithm provided in the embodiment of the present invention, and as shown in fig. 2, the method may include:
step 1: and loading the processed smell data set, and dividing the data set into a training set and a test set.
Step 2: the distance between the test data and each of the training data is calculated using the manhattan distance.
And step 3: sorting according to the increasing relation of the distance values.
And 4, step 4: and selecting proper K values by a cross validation method, and finding K labels with the shortest distance data.
And 5: and determining the occurrence frequency of the category of the first K points, and returning the category with the highest occurrence frequency in the first K points as the prediction classification of the test data.
In conclusion, the invention establishes the corresponding standard odor molecule database by collecting the chemical molecule structure information data in the big data platform, and combines the standard odor molecule database with the chemical theory to preprocess the data and extract the proper characteristics. According to different chemical structure characteristics of different odor substances, similarity classification is carried out on the different odor substances by adopting algorithms such as KNN (Konnen) and BP (Back propagation) neural networks and the like to establish a model, and the recognition rate of the odor is compared, so that the odor of the substance is better recognized through the chemical structure of the substance. When the classification recognition rate is higher, the extracted features can effectively distinguish the substance odor information, namely, the extracted features are closer to the essence of the substance odor and can better represent the odor information of the substance. The method is stable and simple, can provide different thought references for odor identification, can save a great deal of energy and investment required by searching specific odor molecules, and lays a foundation for further researching the odor universality characterization mode of the substance.
According to the invention, by acquiring the chemical molecular structure characteristic data in a big data platform, the PCA algorithm is adopted to reduce the dimension of the original odor information, and the odor is identified by establishing a model through algorithms such as KNN and BP neural networks. Therefore, the big data technology is combined with the chemical correlation theory, and the customized smell molecule database is utilized, so that the storage and subsequent use of smell data are greatly facilitated; the method for establishing the model by using the data mining and machine learning algorithm according to the chemical molecular structure information as the characteristics realizes odor identification from different angles, reduces the deviation between manual measurement and electronic nose measurement, improves the accuracy of odor identification, and expands the research situation that most of the current olfactory field relies on the electronic nose to collect odor data. The model can save a great deal of energy and investment required for searching specific odor molecules, and also lays a foundation for further researching the odor universality characterization mode of the substance. Meanwhile, in the future, it is possible to identify molecules constituting the odor through a specific odor, and the method has a great development space.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (7)
1. A big data analysis-based odor identification method is characterized by comprising the following steps:
acquiring odor molecular structure characteristic information of different substances on a big data platform through a crawler technology, and selecting the odor molecular structure characteristic information of representative and multidimensional chemical molecules to establish a self-defined odor molecular database;
carrying out corresponding data preprocessing on the odor molecule structure characteristic information of the odor molecule database;
performing feature selection and feature extraction on the odor molecular structure feature information subjected to data preprocessing to extract odor comprehensive features which can most represent different substances;
performing feature selection and feature extraction on the odor molecular structure feature information subjected to data preprocessing to extract odor comprehensive features which can most represent different substances;
and establishing a model based on comprehensive odor characteristics which can best represent different substances so as to predict the odor of the substances through the model.
2. The big data analysis-based odor recognition method as claimed in claim 1, wherein building a model based on the comprehensive odor features that can best characterize different substances to predict the odor of the substances through the model comprises:
taking the extracted comprehensive odor characteristics which can represent different substances most as the input of a model, and carrying out training test by continuously adjusting parameters to obtain an optimized model;
and (4) predicting the odor of the unknown sample on the optimized model.
3. The method for identifying odors based on big data analysis according to claim 1, wherein the characteristic selection and characteristic extraction of odor molecule structure characteristic information after data preprocessing is performed to extract comprehensive odor characteristics most representative of different substances, comprising:
and performing dimensionality reduction treatment on the multi-dimensional odor molecular structure characteristic information by adopting a PCA algorithm in combination with actual chemical correlation knowledge, and extracting odor comprehensive characteristics which can represent different substances most.
4. The big data analysis-based odor recognition method of claim 1, wherein the odor molecular structure characteristic information comprises: molecular weight, atomic number, and electrochemical properties.
5. The big-data-analysis-based scent recognition method of claim 1, wherein the data preprocessing comprises: data cleaning and abnormal data elimination.
6. The big data analysis based scent recognition method of claim 1, wherein modeling to predict scent of a substance comprises:
and (3) establishing a model by adopting an algorithm of the KNN and BP neural networks to predict the odor of the substance.
7. The big data analysis based scent recognition method of claim 6, wherein the algorithm using KNN and BP neural networks to model for predicting scent of the substance comprises:
step S401: loading comprehensive odor characteristics which can represent different substances most as a data set, and dividing the data set into a training set and a testing set;
step S402: calculating a distance value between the test data in the test set and the training data in each training set by using the Manhattan distance;
step S403: sorting according to the increasing relation of the distance values;
step S404: selecting proper K values by a cross validation method, and finding out categories corresponding to the test data with the shortest K distance values;
step S405: and determining the occurrence frequency of the category where the first K points are located, and returning the category with the highest occurrence frequency in the first K test data as the prediction classification of the test data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110414388.XA CN113113089A (en) | 2021-04-16 | 2021-04-16 | Smell identification method based on big data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110414388.XA CN113113089A (en) | 2021-04-16 | 2021-04-16 | Smell identification method based on big data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113113089A true CN113113089A (en) | 2021-07-13 |
Family
ID=76718200
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110414388.XA Pending CN113113089A (en) | 2021-04-16 | 2021-04-16 | Smell identification method based on big data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113113089A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110411955A (en) * | 2019-07-15 | 2019-11-05 | 中山大学中山眼科中心 | A kind of artificial intelligence training system based on characterization of molecules predicting of substance color smell |
| CN112580749A (en) * | 2020-12-29 | 2021-03-30 | 上海应用技术大学 | Intelligent fire detection method based on machine olfaction technology |
-
2021
- 2021-04-16 CN CN202110414388.XA patent/CN113113089A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110411955A (en) * | 2019-07-15 | 2019-11-05 | 中山大学中山眼科中心 | A kind of artificial intelligence training system based on characterization of molecules predicting of substance color smell |
| CN112580749A (en) * | 2020-12-29 | 2021-03-30 | 上海应用技术大学 | Intelligent fire detection method based on machine olfaction technology |
Non-Patent Citations (1)
| Title |
|---|
| 马东旭;骆德汉;徐勋庭;冯扬;钟平忠;方桦;: "仿生嗅觉在空气净化器气味识别中的应用研究", 信息技术与网络安全, no. 12 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105928901B (en) | A kind of near-infrared quantitative model construction method that qualitative, quantitative combines | |
| CN103544392B (en) | Medical science Gas Distinguishing Method based on degree of depth study | |
| EP3699582A1 (en) | Gas sensing device and method for operating a gas sensing device | |
| CN111177655B (en) | Data processing method and device and electronic equipment | |
| CN108875118B (en) | Method and device for evaluating accuracy of prediction model of silicon content of blast furnace molten iron | |
| CN116646030B (en) | Tobacco tar component identification method and system based on electronic smoke detection | |
| CN113903409A (en) | Molecular data processing method, model construction and prediction method and related device | |
| CN109669017B (en) | Refinery distillation tower top cut water ion concentration prediction method based on deep learning | |
| CN113255591A (en) | Bearing fault diagnosis method based on random forest and fusion characteristics | |
| CN113343361A (en) | Intelligent monitoring method, device and equipment for vehicle body size and storage medium | |
| Sadoughi et al. | A deep learning approach for failure prognostics of rolling element bearings | |
| CN117390591B (en) | Operation monitoring method and system for coal conveying belt sampling machine based on electric parameter analysis | |
| CN109145685A (en) | Fruits and vegetables EO-1 hyperion quality detecting method based on integrated study | |
| CN117562311A (en) | Detection system of high-performance electronic cigarette atomizer | |
| CN117235661B (en) | AI-based direct drinking water quality monitoring method | |
| CN113113089A (en) | Smell identification method based on big data analysis | |
| Fadchar et al. | Design and development of a neural network—based coconut maturity detector using sound signatures | |
| CN109840386B (en) | Damage identification method based on factor analysis | |
| CN111646332A (en) | Method and system for identifying abnormal operation of elevator | |
| CN113297412A (en) | Music recommendation method and device, electronic equipment and storage medium | |
| Eo et al. | Deep learning framework with essential pre-processing techniques for improving mixed-gas concentration prediction | |
| JPH07334479A (en) | Pattern sorting device and state monitoring system utilizing the device | |
| CN109614714A (en) | A kind of method for establishing model of acidified milk fragrance aesthetic quality control | |
| CN116186507A (en) | Feature subset selection method, device and storage medium | |
| CN113143275B (en) | Electroencephalogram fatigue detection method for quantitative evaluation of sample and characteristic quality in combined manner |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |