CN110968672B - False public opinion identification method for food safety based on neural network - Google Patents
False public opinion identification method for food safety based on neural network Download PDFInfo
- Publication number
- CN110968672B CN110968672B CN201911220854.XA CN201911220854A CN110968672B CN 110968672 B CN110968672 B CN 110968672B CN 201911220854 A CN201911220854 A CN 201911220854A CN 110968672 B CN110968672 B CN 110968672B
- Authority
- CN
- China
- Prior art keywords
- food
- news
- neural network
- products
- false
- 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.)
- Active
Links
- 235000013305 food Nutrition 0.000 title claims abstract description 120
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 26
- 238000013145 classification model Methods 0.000 claims abstract description 13
- 238000003062 neural network model Methods 0.000 claims abstract description 7
- 239000000047 product Substances 0.000 claims description 29
- 238000012549 training Methods 0.000 claims description 25
- 238000013527 convolutional neural network Methods 0.000 claims description 13
- 238000002372 labelling Methods 0.000 claims description 12
- 230000002457 bidirectional effect Effects 0.000 claims description 8
- 235000013350 formula milk Nutrition 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 7
- 229920002472 Starch Polymers 0.000 claims description 6
- 235000019698 starch Nutrition 0.000 claims description 6
- 239000008107 starch Substances 0.000 claims description 6
- 230000000737 periodic effect Effects 0.000 claims description 5
- WPYMKLBDIGXBTP-UHFFFAOYSA-N benzoic acid Chemical compound OC(=O)C1=CC=CC=C1 WPYMKLBDIGXBTP-UHFFFAOYSA-N 0.000 claims description 4
- 239000004278 EU approved seasoning Substances 0.000 claims description 3
- 244000046052 Phaseolus vulgaris Species 0.000 claims description 3
- 235000010627 Phaseolus vulgaris Nutrition 0.000 claims description 3
- 244000269722 Thea sinensis Species 0.000 claims description 3
- 244000299461 Theobroma cacao Species 0.000 claims description 3
- 235000009470 Theobroma cacao Nutrition 0.000 claims description 3
- 235000013361 beverage Nutrition 0.000 claims description 3
- 235000009508 confectionery Nutrition 0.000 claims description 3
- 235000013365 dairy product Nutrition 0.000 claims description 3
- 235000005911 diet Nutrition 0.000 claims description 3
- 230000000378 dietary effect Effects 0.000 claims description 3
- 239000008157 edible vegetable oil Substances 0.000 claims description 3
- 235000013373 food additive Nutrition 0.000 claims description 3
- 239000002778 food additive Substances 0.000 claims description 3
- 235000011194 food seasoning agent Nutrition 0.000 claims description 3
- 235000013569 fruit product Nutrition 0.000 claims description 3
- 235000013402 health food Nutrition 0.000 claims description 3
- 235000013622 meat product Nutrition 0.000 claims description 3
- 235000013606 potato chips Nutrition 0.000 claims description 3
- 235000013311 vegetables Nutrition 0.000 claims description 3
- 235000014101 wine Nutrition 0.000 claims description 3
- 239000005711 Benzoic acid Substances 0.000 claims description 2
- IOVCWXUNBOPUCH-UHFFFAOYSA-M Nitrite anion Chemical compound [O-]N=O IOVCWXUNBOPUCH-UHFFFAOYSA-M 0.000 claims description 2
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 235000010233 benzoic acid Nutrition 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- OIQPTROHQCGFEF-UHFFFAOYSA-L chembl1371409 Chemical compound [Na+].[Na+].OC1=CC=C2C=C(S([O-])(=O)=O)C=CC2=C1N=NC1=CC=C(S([O-])(=O)=O)C=C1 OIQPTROHQCGFEF-UHFFFAOYSA-L 0.000 claims description 2
- 239000000126 substance Substances 0.000 claims description 2
- 235000015895 biscuits Nutrition 0.000 claims 1
- 239000004519 grease Substances 0.000 claims 1
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 235000014510 cooky Nutrition 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 239000003925 fat Substances 0.000 description 2
- 239000003921 oil Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
Images
Classifications
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a food safety false public opinion identification method based on a neural network, relates to the field of artificial intelligence, and can monitor network public opinions and screen out false news. The method comprises the following steps: building a food risk factor entity library; building a food name entity library; constructing a dynamic official news rumor-breaking library; building a real and false news neural network classification model; inputting latest news public sentiments, marking the food names and the risk factors involved in the news by comparing and searching in the food risk factor entity library and the food name entity library, preliminarily classifying the news, then carrying out similarity comparison according to the official news rumor splitting library, and if the related rumor splitting news is not searched, then carrying out true and false news classification by using a neural network model.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a false public opinion identification method for food safety based on a neural network.
Background
At present, a large amount of news public opinion reports related to food safety are generated every day on the network, if the information is left alone and is not supervised, some false news can cause unnecessary social panic and disturb social order, so that the production life of people is influenced, the research and the development of safety monitoring and early warning comprehensive information of key varieties of food safety supervision can be realized, the collection, the analysis and the information distribution of data information such as food safety supervision key variety supervision spot check, risk monitoring and the like can be realized, the supervision efficiency is greatly improved, the targets of deep mining, full utilization and information sharing of food safety information can be achieved, beneficial experience reference is provided for the food safety supervision informatization construction, and a powerful tool is provided for food safety supervision departments and consumers to avoid food harm. In the supervision engineering, the method can rapidly conduct public opinion on news and rumors, and has great research significance in applying natural language processing technology to help conduct food safety supervision.
Disclosure of Invention
The invention provides a method for classifying news texts in order to screen out false news related to food safety on the network, which aims to meet the requirement of monitoring news related to food safety on the Internet at present.
The method provided by the invention comprises the following steps: a false public opinion identification method based on neural network for food safety comprises the following steps:
step 1, constructing a bottom knowledge base which comprises a food name database and a food risk factor database;
step 2, constructing an official news rumor splitting library updated in real time;
step 3, constructing a method of combining a bidirectional long-and-short-term memory network and a conditional random field as a neural network model for entity recognition, and training by using news corpora with correctly marked entity names and risk factor names to obtain a final model; during prediction, vector representation of a news corpus sequence is used as input, and a labeling result of the sequence is obtained and used as output; obtaining the food name and the risk factor name related in the news according to the labeling result, judging whether the labeling result is reliable or not by combining a bottom knowledge base, and preliminarily determining the truth of the news;
And 4, constructing a convolutional neural network model as a news public opinion classification model, training by using news corpora which are correctly labeled with true and false to obtain the convolutional neural network classification model, and inputting the labeling result in the step 3 when predicting by using a neural network to finally obtain true and false classification of the news public opinion as output.
Further, the food name database includes names of existing foods, including food processed products, edible oils, fats and oils and products thereof, seasonings, meat products, dairy products, beverages, instant foods, cookies, cans, frozen drinks, quick-frozen foods, potato chips and puffed foods, confectionery products, tea and related products, wines, vegetable products, fruit products, roasted and nut products, egg products, cocoa and roasted coffee products, sugar, aquatic products, starch and starch products, cakes, bean products, bee products, health foods, special dietary foods, special medical use formula foods, infant formula foods, catering foods, edible agricultural products, food additives; for some newly generated foods, the food item name database is replenished in time by periodic updates.
Further, the food risk factor refers to a substance whose content is required to be monitored to be excessive in food, and the names of the food risk factors in the food risk factor database include: lead, benzoic acid, nitrite, sunset yellow, total colony count and different food risk factor names corresponding to different foods; for some newly defined or newly discovered risk factors, the air risk factor database is timely supplemented by periodic updates.
Further, a two-way long-and-short-term memory network and a conditional random field model are trained to serve as an entity recognition model capable of accurately recognizing food names and food risk factor names.
Further, in the step 1, after the food name database and the food risk factor database are constructed, the direct relationship between the two underlying knowledge bases is connected, the food factors to be detected of each food are associated, and the maximum content value is recorded.
Further, in step 3, the news public sentiment is labeled, the labeled content comprises a food name and a food risk factor name related to the news public sentiment, and the truth of the news public sentiment is preliminarily judged according to the labeling result and an official news rumor library.
Further, training a convolutional neural network model to serve as a classification model of news public sentiment, converting news corpora into vector representation to serve as input of a neural network, training the neural network classification model, building a model by using the convolutional neural network, and training the existing news corpora to obtain a final model; the final output result of the model is true and false of news public sentiment.
Further, training two neural network models, including a bidirectional long-time and short-time memory network model and a conditional random field model for identifying the food entity name and the risk factor name, and a convolutional neural network model for news and public opinion classification; when training is started, the weight is initialized randomly, after the last layer of result is obtained through calculation of a neural network, the cross entropy between a predicted value and a true value is calculated to serve as a loss function, the loss function is minimized through an adaptive moment estimation algorithm, and the learning rate is adjusted according to the training process; in the training process, in order to improve the training efficiency, one batch of data is input every time, and meanwhile, in order to prevent overfitting, a certain proportion of weight values are randomly set to be 0 in the training process.
The method has the advantages that the truth of news public sentiment can be judged quickly and efficiently, main food names and entity risk factor names involved in news are marked clearly, analysis of the whole news event is shown clearly, and a monitor is assisted to make a correct decision.
Drawings
FIG. 1 is a schematic flow chart of a neural network-based food safety false public opinion identification method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of the dynamic update of the underlying knowledge base;
FIG. 3 is a network diagram of entity identification;
FIG. 4 is a schematic diagram of a convolutional neural network classification.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments, and all other embodiments obtained by a person skilled in the art based on the embodiments of the present invention belong to the protection scope of the present invention without creative efforts.
According to an embodiment of the invention, the invention provides a false public opinion identification method for food safety based on a neural network, which comprises the following steps:
step 1, constructing a bottom knowledge base which comprises a food name database and a food risk factor database;
step 2, constructing an official news rumor splitting library updated in real time;
step 3, constructing a method of combining a bidirectional long-and-short-term memory network and a conditional random field as a neural network model for entity recognition, and training by using news corpora with correctly marked entity names and risk factor names to obtain a final model; during prediction, vector representation of a news corpus sequence is used as input, and a labeling result of the sequence is obtained and used as output; obtaining the food name and the risk factor name related in the news according to the labeling result, judging whether the labeling result is reliable or not by combining a bottom knowledge base, and preliminarily determining the truth of the news;
And 4, constructing a convolutional neural network model as a news public opinion classification model, training by using news corpora which are correctly labeled with true and false to obtain the convolutional neural network classification model, and finally obtaining true and false classification of news public opinions as output by using the labeling result in the step 3 as input when the neural network is used for prediction.
Further, the food name database includes names of existing foods, including food processed products, edible oils, fats and oils and products thereof, seasonings, meat products, dairy products, beverages, instant foods, cookies, cans, frozen drinks, quick-frozen foods, potato chips and puffed foods, confectionery products, tea and related products, wines, vegetable products, fruit products, roasted and nut products, egg products, cocoa and roasted coffee products, sugar, aquatic products, starch and starch products, cakes, bean products, bee products, health foods, special dietary foods, special medical use formula foods, infant formula foods, catering foods, edible agricultural products, food additives; for some newly generated foods, the food item name database is replenished in time by periodic updates.
Referring to fig. 1, showing an overall schematic diagram of the method provided by the invention, a news public opinion text is used as input, entity names involved in the text are marked through entity identification, a self-constructed bottom knowledge base is provided, the bottom knowledge base comprises a food name knowledge base and a food risk factor knowledge base, matching is carried out according to the bottom knowledge base, evidence support is obtained from an official rumor library according to matched information, a judgment result is directly given for a text with sufficient evidence, a text with insufficient evidence is converted into vector representation and input into a neural network classification model, and finally the network gives true and false news as a final output result.
In the embodiment shown in fig. 2, the bottom knowledge base includes a food name knowledge base and a food risk factor base, and the food name knowledge base is constructed by the invention and stores the basic attributes of food names, short names and the like; the food risk factor library comprises the names of risk factors to be detected in food, the names of the corresponding food, the maximum content in detection, the detection method and other attributes, and because the conditions of the same risk factor to be detected exist among different foods, the judgment conditions of the risk factors in different foods are different. In addition, the underlying knowledge base needs to be dynamically updated, the method has real-time or regular updating for some newly generated foods or newly discovered food risk factors, and finally determines whether to dynamically update the underlying knowledge base or not by comparing information in the official public rumor library for food names and food risk factor names which are not matched from the underlying entity library.
In another embodiment shown in fig. 3, a news text is converted into a vector representation as an input of a bidirectional long-short memory network, and the bidirectional long-short memory network and a conditional random field are combined to obtain an entity name contained in the text, so that a question Q (w) with the input length of n is provided1,w2,…,wn) Wherein w isiRepresents the ith word; using one-hot encoding to obtain a vector representation, X (X), for each word1,x2,…,xn)，xiA vector representation representing the ith word;then inputting X into two different long-time memory networks in a positive sequence and a reverse sequence respectively, and finally obtaining a state h at a time ttContaining the context information at the moment. The output of the bidirectional long-time memory network layer is each Chinese character wtThe probability marked as each label and the final probability matrix are used as the input of the conditional random field layer to calculate the scores of different label sequences, and this way, unreasonable label sequences, such as 'B-ER, O, I-ER' and the like, can be effectively avoided.
In the embodiment shown in fig. 4, for a news text which does not have enough evidence to prove the truth of the news text, it needs to be judged through a neural network model, the processed news text is vectorized and represented, and finally, through n layers of convolutional layers and pooling layers, the probability of each final classification of the news is obtained through a softmax function through a full connection layer, and the formula of the softmax function is as follows: 
The function maps the output of the neuron into the interval (0,1), where n represents the number of classes, i represents a class in j, and giA value, P(s), representing the classificationi) Representing the probability of the ith class.
Although illustrative embodiments of the present invention have been described above to facilitate the understanding of the present invention by those skilled in the art, it should be understood that the present invention is not limited to the scope of the embodiments, but various changes may be apparent to those skilled in the art, and it is intended that all inventive concepts utilizing the inventive concepts set forth herein be protected without departing from the spirit and scope of the present invention as defined and limited by the appended claims.
Claims (7)
1. A false public opinion identification method of food safety based on neural network is characterized by comprising the following steps:
step 1, constructing a bottom knowledge base which comprises a food name database and a food risk factor database;
step 2, constructing an official news rumor splitting library updated in real time;
step 3, constructing a method of combining a bidirectional long-and-short-term memory network and a conditional random field as a neural network model for entity recognition, and training by using news corpora with correctly marked entity names and risk factor names to obtain a final model; during prediction, vector representation of a news corpus sequence is used as input, entity names related in a text are marked through entity identification, matching is carried out according to a bottom knowledge base, and then a marking result of the sequence is obtained according to matched information and is used as output; obtaining the food name and the risk factor name related in the news according to the labeling result, judging whether the labeling result is reliable or not by combining a bottom knowledge base, and preliminarily determining the truth of the news; obtaining evidence support from an official rumor-seeking library, directly giving a judgment result for a text with sufficient evidence, converting the text with insufficient evidence into vector representation, and inputting the vector representation into a neural network classification model;
Step 4, constructing a convolutional neural network model as a news public opinion classification model, training by using news corpora which are correctly labeled with true and false to obtain the convolutional neural network classification model, and when the neural network is used for prediction, taking the labeling result in the step 3 as input, and finally obtaining true and false classification of news public opinion as output;
further comprising: training two neural network models, including a two-way long-time memory network model and a conditional random field model for recognizing food entity names and risk factor names, and a convolutional neural network model for news public opinion classification; when training is started, the weight is initialized randomly, after the last layer of result is obtained through calculation of a neural network, the cross entropy between a predicted value and a true value is calculated to serve as a loss function, the loss function is minimized through an adaptive moment estimation algorithm, and the learning rate is adjusted according to the training process; in the training process, in order to improve the training efficiency, one batch of data is input every time, and meanwhile, in order to prevent overfitting, a certain proportion of weight values are randomly set to be 0 in the training process.
2. The false public opinion recognition method based on neural network for food safety as claimed in claim 1, wherein the false public opinion recognition method comprises:
The food name database comprises the names of existing foods, including food processing products, edible oil, grease and products thereof, seasonings, meat products, dairy products, beverages, instant foods, biscuits, cans, frozen drinks, quick-frozen foods, potato chips and puffed foods, candy products, tea leaves and related products, wines, vegetable products, fruit products, roasted food and nut products, egg products, cocoa and roasted coffee products, sugar, aquatic products, starch and starch products, cakes, bean products, bee products, health foods, special dietary foods, special medical purpose formula foods, infant formula foods, catering foods, edible agricultural products and food additives; for some newly generated foods, the food item name database is replenished in time by periodic updates.
3. The false public opinion recognition method based on neural network for food safety as claimed in claim 1, wherein the false public opinion recognition method comprises:
the food risk factor refers to a substance with the content exceeding the standard or not, and the food risk factor name in the food risk factor database comprises the following components: lead, benzoic acid, nitrite, sunset yellow, the total number of colonies, and the names of food risk factors corresponding to different foods are different; for some newly defined or newly discovered risk factors, the risk factor database is replenished in time by periodic updates.
4. The false public opinion recognition method for food safety based on neural network as claimed in claim 1, wherein:
and training a bidirectional long-time memory network and a conditional random field model as an entity recognition model capable of accurately recognizing food names and food risk factor names.
5. The false public opinion recognition method for food safety based on neural network as claimed in claim 1, wherein:
in the step 1, after the food name database and the food risk factor database are constructed, the direct relation between the two bottom knowledge bases is connected, the food factors to be detected of each food are associated, and the value of the maximum content is recorded.
6. The false public opinion recognition method for food safety based on neural network as claimed in claim 1, wherein:
and step 3, marking the news public sentiment, wherein the marked content comprises a food name and a food risk factor name related to the news public sentiment, and preliminarily judging the truth and falseness of the news public sentiment according to a marking result and an official news rumor library.
7. The false public opinion recognition method for food safety based on neural network as claimed in claim 1, wherein:
Training a convolutional neural network model as a classification model of news public sentiment truth and falseness, converting news corpora into vector representation as input of a neural network, training the neural network classification model, building a model by using the convolutional neural network, and training through the existing news corpora to obtain a final model; the final output result of the model is the truth and falseness of the news public opinion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911220854.XA CN110968672B (en) | 2019-12-03 | 2019-12-03 | False public opinion identification method for food safety based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911220854.XA CN110968672B (en) | 2019-12-03 | 2019-12-03 | False public opinion identification method for food safety based on neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110968672A CN110968672A (en) | 2020-04-07 |
| CN110968672B true CN110968672B (en) | 2022-06-10 |
Family
ID=70032724
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911220854.XA Active CN110968672B (en) | 2019-12-03 | 2019-12-03 | False public opinion identification method for food safety based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110968672B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112527956A (en) * | 2020-12-08 | 2021-03-19 | 北京工商大学 | Food safety public opinion event extraction method based on deep learning |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102929918A (en) * | 2012-09-20 | 2013-02-13 | 西北工业大学 | False online public opinion identification method |
| CN110210016A (en) * | 2019-04-25 | 2019-09-06 | 中国科学院计算技术研究所 | Bilinearity neural network Deceptive news detection method and system based on style guidance |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160335534A1 (en) * | 2015-05-14 | 2016-11-17 | Thalchemy Corporation | Neural sensor hub system |
-
2019
- 2019-12-03 CN CN201911220854.XA patent/CN110968672B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102929918A (en) * | 2012-09-20 | 2013-02-13 | 西北工业大学 | False online public opinion identification method |
| CN110210016A (en) * | 2019-04-25 | 2019-09-06 | 中国科学院计算技术研究所 | Bilinearity neural network Deceptive news detection method and system based on style guidance |
Non-Patent Citations (1)
| Title |
|---|
| "基于深度主动学习的信息安全领域命名实体识别研究";彭嘉毅等;《四川大学学报(自然科学版)》;20190531;87-92页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110968672A (en) | 2020-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106383815B (en) | In conjunction with the neural network sentiment analysis method of user and product information | |
| CN111879772B (en) | Food safety intelligent management method and system based on big data | |
| CN107851097A (en) | Data analysis system, data analysing method, data analysis program and storage medium | |
| CA2825742A1 (en) | Marketing device, marketing method, program and recording medium | |
| WO2020211756A1 (en) | Semantic recognition method, device, computer readable storage medium and computer apparatus | |
| CN111564199A (en) | Intelligent nutrition intervention method and terminal | |
| CN110968672B (en) | False public opinion identification method for food safety based on neural network | |
| CN106446191B (en) | A kind of multiple features network flow row label prediction technique returned based on Logistic | |
| CN109636440B (en) | Commodity satisfaction evaluation method and device, storage medium and electronic equipment | |
| CN111755098A (en) | Balance diet mathematical model establishing method | |
| Barthwal et al. | New trends in the development and application of artificial intelligence in food processing | |
| Pratiwi et al. | Classification of Orange Fruit Using Convolutional Neural Network, Support Vector Machine, K-Nearest Neighbor and Naive Bayes Methods Based on Color Analysis | |
| CN109657123A (en) | A kind of food safety affair clustering method based on comentropy | |
| WO2024002216A1 (en) | Service problem attribution method and device | |
| Hidayat et al. | Decision support system for selection of staples food and food commodity price prediction post-covid-19 using simple additive weighting and multiple linear regression methods | |
| Yeruva et al. | Exploring social contextual influences on healthy eating using big data analytics | |
| CN115375408A (en) | Article information recommendation method and device and electronic equipment | |
| CN116205688A (en) | Fresh product information processing method and device, computer equipment and storage medium | |
| CN116340383A (en) | Searching method, device, medium and equipment based on correlation of query and result | |
| Tiwari et al. | Artificial Intelligence in Food Processing | |
| CN113052497A (en) | Criminal worker risk prediction method based on dynamic and static feature fusion learning | |
| White et al. | Investigating eating behaviours using topic models | |
| Reka et al. | Deep Learning-Based Classification of Rotten Fruits and Identification of Shelf Life. | |
| Siddique | IMPLEMENTING BIG DATA ANALYTICS APPROACHES TO IMPROVE FOOD QUALITY AND MINIMIZE FOOD WASTE AND LOSS | |
| AU2021106125A4 (en) | A glycemic load calculator with image detection |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |