Modified National Institute of Standards and Technology database
E363685
The Modified National Institute of Standards and Technology database is a large, standardized collection of handwritten digit images widely used for training and evaluating image processing and machine learning algorithms.
All labels observed (1)
| Label | Occurrences |
|---|---|
| Modified National Institute of Standards and Technology database canonical | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T3507174 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Modified National Institute of Standards and Technology database Context triple: [MNIST, fullName, Modified National Institute of Standards and Technology database]
-
A.
ISO 10383 MIC database
The ISO 10383 MIC database is an international registry that assigns and maintains standardized Market Identifier Codes (MICs) for securities and derivatives trading venues and related entities worldwide.
-
B.
Integrated Automated Fingerprint Identification System
The Integrated Automated Fingerprint Identification System is the FBI’s large-scale computerized system for storing, searching, and matching fingerprint and biometric data to support criminal identification and investigative work.
-
C.
Combined DNA Index System
The Combined DNA Index System (CODIS) is a national DNA database in the United States that enables law enforcement agencies to store, compare, and match DNA profiles for criminal investigations and identification.
-
D.
NISTAR
NISTAR is a radiometer aboard the Deep Space Climate Observatory that measures Earth’s reflected and emitted radiation to study the planet’s energy budget and climate.
-
E.
National Instant Criminal Background Check System
The National Instant Criminal Background Check System (NICS) is a U.S. system used to quickly determine whether a prospective firearm buyer is legally eligible to purchase guns.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
NED2
Entity disambiguation (via description)
gpt-5-mini-2025-08-07
Target entity: Modified National Institute of Standards and Technology database Target entity description: The Modified National Institute of Standards and Technology database is a large, standardized collection of handwritten digit images widely used for training and evaluating image processing and machine learning algorithms.
-
A.
ISO 10383 MIC database
The ISO 10383 MIC database is an international registry that assigns and maintains standardized Market Identifier Codes (MICs) for securities and derivatives trading venues and related entities worldwide.
-
B.
Integrated Automated Fingerprint Identification System
The Integrated Automated Fingerprint Identification System is the FBI’s large-scale computerized system for storing, searching, and matching fingerprint and biometric data to support criminal identification and investigative work.
-
C.
Combined DNA Index System
The Combined DNA Index System (CODIS) is a national DNA database in the United States that enables law enforcement agencies to store, compare, and match DNA profiles for criminal investigations and identification.
-
D.
NISTAR
NISTAR is a radiometer aboard the Deep Space Climate Observatory that measures Earth’s reflected and emitted radiation to study the planet’s energy budget and climate.
-
E.
National Instant Criminal Background Check System
The National Instant Criminal Background Check System (NICS) is a U.S. system used to quickly determine whether a prospective firearm buyer is legally eligible to purchase guns.
- F. None of above. chosen
Statements (49)
| Predicate | Object |
|---|---|
| instanceOf |
benchmark dataset
ⓘ
handwritten digit dataset ⓘ image dataset ⓘ machine learning dataset ⓘ |
| abbreviation | MNIST ⓘ |
| basedOn |
NIST Special Database 1
ⓘ
NIST Special Database 3 ⓘ |
| benchmarkStatus | de facto standard for handwritten digit recognition ⓘ |
| classLabels | digits 0–9 ⓘ |
| contains | grayscale images of handwritten digits ⓘ |
| creator |
Christopher J. C. Burges
ⓘ
Corinna Cortes ⓘ Yann LeCun ⓘ |
| dataSplit |
test set
ⓘ
training set ⓘ |
| dataType | unsigned 8-bit integer pixel values ⓘ |
| difficultyLevel | relatively easy classification task ⓘ |
| domain | handwritten digit recognition ⓘ |
| fileFormat | IDX ⓘ |
| hostedAt | Yann LeCun’s NYU web pages ⓘ |
| imageChannels | 1 ⓘ |
| imageColorSpace | grayscale ⓘ |
| imageHeight | 28 pixels ⓘ |
| imageWidth | 28 pixels ⓘ |
| influenced |
EMNIST dataset
ⓘ
Fashion-MNIST ⓘ
surface form:
Fashion-MNIST dataset
KMNIST ⓘ
surface form:
KMNIST dataset
|
| license | freely available for research and educational use ⓘ |
| notableFor |
standardized train-test split
ⓘ
widespread use in deep learning research ⓘ |
| numberOfClasses | 10 ⓘ |
| pixelValueRange | 0–255 ⓘ |
| preprocessing |
centered in fixed-size image
ⓘ
size-normalized digits ⓘ |
| publisher | National Institute of Standards and Technology ⓘ |
| subjectArea |
computer vision
ⓘ
machine learning education ⓘ pattern recognition ⓘ |
| testSetSize | 10000 ⓘ |
| timePeriod | 1990s ⓘ |
| totalNumberOfImages | 70000 ⓘ |
| trainingSetSize | 60000 ⓘ |
| typicalInputShape | 28x28 ⓘ |
| typicalUse |
convolutional neural network training
ⓘ
supervised learning ⓘ |
| usedFor |
benchmarking machine learning algorithms
ⓘ
evaluating image processing algorithms ⓘ optical character recognition research ⓘ training image classification models ⓘ |
How these facts were elicited
The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.
Instruction
You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10. # Requirements - If you don't know the subject at all, return an empty list. - If the subject is not a named entity, return an empty list. - Include at least one triple where predicate is "instanceOf". - Do not get too wordy. - Separate several objects into multiple triples with one object.
Input
Subject: Modified National Institute of Standards and Technology database Description of subject: The Modified National Institute of Standards and Technology database is a large, standardized collection of handwritten digit images widely used for training and evaluating image processing and machine learning algorithms.
Referenced by (1)
Full triples — surface form annotated when it differs from this entity's canonical label.