Human and machine learning when related to language: Difference between revisions
Mr. MacKenty (talk | contribs) No edit summary |
Mr. MacKenty (talk | contribs) No edit summary |
||
(12 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
[[file:simulation.png|right|frame|HL content: Modeling & Simulation<ref>http://www.flaticon.com/</ref>]] | [[file:simulation.png|right|frame|HL content: Modeling & Simulation<ref>http://www.flaticon.com/</ref>]] | ||
== From the IB == | == From the IB == | ||
Using knowledge, such as the syntax of a language, leads to an appreciation of the difficulties involved in machine language learning. Students should be familiar with the concept of cognitive learning and the use of heuristics and probabilities in machine learning. | Using knowledge, such as the syntax of a language, leads to an appreciation of the difficulties involved in machine language learning. Students should be familiar with the concept of cognitive learning and the use of heuristics and probabilities in machine learning. | ||
== Language == | |||
Linguists distinguish between language acquisition and language learning. | |||
Children acquire language through a subconscious process during which they are unaware of grammatical rules. This happens especially when they acquire their first language. They repeat what is said to them and get a feel for what is and what is not correct. In order to acquire a language, they need a source of natural communication, which is usually the mother, the father, or the caregiver. | |||
Language learning, on the other hand, is the result of direct instruction in the rules of language. Language learning is not an age-appropriate activity for very young children as learning presupposes that learners have a conscious knowledge of the new language and can talk about that knowledge. They usually have a basic knowledge of the grammar.<ref>http://www.utesinternationallounge.com/language-acquisition-versus-language-learning/</ref> | |||
=== How children learn language === | |||
==== An overview ==== | |||
Children learn language by observing their environment, listening to the people around them, and connecting the dots between what they see and hear. Among other things, this helps children establish their language’s word order, such as where subjects and verbs fall in a sentence.<ref>http://news.mit.edu/2018/machines-learn-language-human-interaction-1031</ref> | |||
A more in-depth discussion, used with gratitude from [https://www.verywellfamily.com/how-do-children-learn-language-1449116 this resource] : | |||
==== Stage One: Learning Sounds ==== | |||
When babies are born, they can make and hear all the sounds in all the languages in the world. That’s about 150 sounds in about 6500 languages, though no language uses all of those sounds. The sounds a language uses are called phonemes and English has about 44. Some languages use more and some use fewer. | |||
In this stage, babies learn which phonemes belong to the language they are learning and which don’t. The ability to recognize and produce those sounds is called “phonemic awareness,” which is important for children learning to read. | |||
==== Stage Two: Learning Words ==== | |||
At this stage, children essentially learn how the sounds in a language go together to make meaning. For example, they learn that the sounds m-ah-m-ee refer to the “being” who cuddles and feeds them, their mommy. | |||
This is a significant step because everything we say is really just a stream of sounds. To make sense of those sounds, a child must be able to recognize where one word ends and another one begins. These are called “word boundaries.” | |||
However, children are not learning words, exactly. They are actually learning morphemes, which may or may not be words. A morpheme is simply a sound or sounds that have a meaning, like a word mommy. | |||
The word mommies, however, has two morphemes: mommy and –s. Children at this stage can recognize that the –s means "more than one." They also begin to associate that meaning with other words when the sound is added. | |||
==== Stage Three: Learning Sentences ==== | |||
During this stage, children learn how to create sentences. That means they can put words in the correct order. For example, they learn that in English we say "I want a cookie" and "I want a chocolate cookie," not "Want I a cookie" or "I want cookie chocolate." | |||
Children also learn the difference between grammatical correctness and meaning. Noam Chomsky created an example of this difference in the sentence “Colorless green ideas sleep furiously.” Children will know that although the sentence is grammatically correct, it doesn’t make sense. They know that green is a color and can't, therefore, be colorless. | |||
=== How computers learn language === | |||
Up to the 1980s, most natural language processing systems were based on complex sets of hand-written rules. However, this is rarely robust to natural language variation. Starting in the late 1980s, however, there was a revolution in natural language processing with the introduction of machine learning algorithms for language processing <ref>https://en.wikipedia.org/wiki/Natural_language_processing</ref> | |||
The machine-learning paradigm calls instead for using statistical inference (statistical inference is the process of using data analysis to deduce properties of an underlying probability distribution<ref>https://en.wikipedia.org/wiki/Statistical_inference</ref>) to automatically learn such rules through the analysis of large corpora of typical real-world examples (a corpus (plural, "corpora") is a set of documents, possibly with human or computer annotations).<ref>https://en.wikipedia.org/wiki/Natural_language_processing</ref> | |||
Please understand some the areas where computer decompose a language. As part of this article, please review the following wikipedia articles: | |||
* [https://en.wikipedia.org/wiki/Natural_language_processing#Syntax Syntax] | |||
* [https://en.wikipedia.org/wiki/Natural_language_processing#Semantics Semantics] | |||
* [https://en.wikipedia.org/wiki/Natural_language_processing#Discourse Discourse] | |||
* [https://en.wikipedia.org/wiki/Natural_language_processing#Speech Speech] | |||
Also, these videos may help you understand: | |||
* https://www.youtube.com/watch?v=_GdSC1Z1Kzs | |||
* [https://www.youtube.com/watch?v=fOvTtapxa9c As usual, crash course helps us understand this] | |||
== Heuristic == | == Heuristic == | ||
Line 9: | Line 67: | ||
In computer science, artificial intelligence, and mathematical optimization, a heuristic is a technique designed for solving a problem more quickly when classic methods are too slow, or for finding an approximate solution when classic methods fail to find any exact solution. This is achieved by trading optimality, completeness, accuracy, or precision for speed. In a way, it can be considered a shortcut.<ref>https://en.wikipedia.org/wiki/Heuristic_(computer_science)</ref> | In computer science, artificial intelligence, and mathematical optimization, a heuristic is a technique designed for solving a problem more quickly when classic methods are too slow, or for finding an approximate solution when classic methods fail to find any exact solution. This is achieved by trading optimality, completeness, accuracy, or precision for speed. In a way, it can be considered a shortcut.<ref>https://en.wikipedia.org/wiki/Heuristic_(computer_science)</ref> | ||
== The difference == | |||
Human cognitive learning is a complex thought process which includes past experience of the ways in which words are used and ambient senses such as facial expressions when words are used etc; | |||
But machine learning involves following preset rules combined with heuristics – building up experience over time. Using probabilities to place words/phrases correctly; | |||
These different approaches to learning make it difficult for machines to use/interpret language in the same way as humans | |||
Research into cognitive learning/human thinking could mean that (given time), machines could be taught natural language | |||
== Learn more == | |||
* [https://en.wikipedia.org/wiki/Language_processing_in_the_brain This is a good wikipedia article which more deeply explains how humans process language] | |||
* [https://en.wikipedia.org/wiki/Natural_language_processing This is a good wikipedia article which more deeply explains how computers process language] | |||
== Standards == | == Standards == |
Latest revision as of 15:01, 1 March 2019
![](/images/5/5a/Simulation.png)
From the IB[edit]
Using knowledge, such as the syntax of a language, leads to an appreciation of the difficulties involved in machine language learning. Students should be familiar with the concept of cognitive learning and the use of heuristics and probabilities in machine learning.
Language[edit]
Linguists distinguish between language acquisition and language learning. Children acquire language through a subconscious process during which they are unaware of grammatical rules. This happens especially when they acquire their first language. They repeat what is said to them and get a feel for what is and what is not correct. In order to acquire a language, they need a source of natural communication, which is usually the mother, the father, or the caregiver.
Language learning, on the other hand, is the result of direct instruction in the rules of language. Language learning is not an age-appropriate activity for very young children as learning presupposes that learners have a conscious knowledge of the new language and can talk about that knowledge. They usually have a basic knowledge of the grammar.[2]
How children learn language[edit]
An overview[edit]
Children learn language by observing their environment, listening to the people around them, and connecting the dots between what they see and hear. Among other things, this helps children establish their language’s word order, such as where subjects and verbs fall in a sentence.[3]
A more in-depth discussion, used with gratitude from this resource :
Stage One: Learning Sounds[edit]
When babies are born, they can make and hear all the sounds in all the languages in the world. That’s about 150 sounds in about 6500 languages, though no language uses all of those sounds. The sounds a language uses are called phonemes and English has about 44. Some languages use more and some use fewer.
In this stage, babies learn which phonemes belong to the language they are learning and which don’t. The ability to recognize and produce those sounds is called “phonemic awareness,” which is important for children learning to read.
Stage Two: Learning Words[edit]
At this stage, children essentially learn how the sounds in a language go together to make meaning. For example, they learn that the sounds m-ah-m-ee refer to the “being” who cuddles and feeds them, their mommy.
This is a significant step because everything we say is really just a stream of sounds. To make sense of those sounds, a child must be able to recognize where one word ends and another one begins. These are called “word boundaries.”
However, children are not learning words, exactly. They are actually learning morphemes, which may or may not be words. A morpheme is simply a sound or sounds that have a meaning, like a word mommy.
The word mommies, however, has two morphemes: mommy and –s. Children at this stage can recognize that the –s means "more than one." They also begin to associate that meaning with other words when the sound is added.
Stage Three: Learning Sentences[edit]
During this stage, children learn how to create sentences. That means they can put words in the correct order. For example, they learn that in English we say "I want a cookie" and "I want a chocolate cookie," not "Want I a cookie" or "I want cookie chocolate."
Children also learn the difference between grammatical correctness and meaning. Noam Chomsky created an example of this difference in the sentence “Colorless green ideas sleep furiously.” Children will know that although the sentence is grammatically correct, it doesn’t make sense. They know that green is a color and can't, therefore, be colorless.
How computers learn language[edit]
Up to the 1980s, most natural language processing systems were based on complex sets of hand-written rules. However, this is rarely robust to natural language variation. Starting in the late 1980s, however, there was a revolution in natural language processing with the introduction of machine learning algorithms for language processing [4]
The machine-learning paradigm calls instead for using statistical inference (statistical inference is the process of using data analysis to deduce properties of an underlying probability distribution[5]) to automatically learn such rules through the analysis of large corpora of typical real-world examples (a corpus (plural, "corpora") is a set of documents, possibly with human or computer annotations).[6]
Please understand some the areas where computer decompose a language. As part of this article, please review the following wikipedia articles:
Also, these videos may help you understand:
Heuristic[edit]
In computer science, artificial intelligence, and mathematical optimization, a heuristic is a technique designed for solving a problem more quickly when classic methods are too slow, or for finding an approximate solution when classic methods fail to find any exact solution. This is achieved by trading optimality, completeness, accuracy, or precision for speed. In a way, it can be considered a shortcut.[7]
The difference[edit]
Human cognitive learning is a complex thought process which includes past experience of the ways in which words are used and ambient senses such as facial expressions when words are used etc; But machine learning involves following preset rules combined with heuristics – building up experience over time. Using probabilities to place words/phrases correctly; These different approaches to learning make it difficult for machines to use/interpret language in the same way as humans Research into cognitive learning/human thinking could mean that (given time), machines could be taught natural language
Learn more[edit]
- This is a good wikipedia article which more deeply explains how humans process language
- This is a good wikipedia article which more deeply explains how computers process language
Standards[edit]
- Discuss the differences between human and machine learning when related to language.
References[edit]
- ↑ http://www.flaticon.com/
- ↑ http://www.utesinternationallounge.com/language-acquisition-versus-language-learning/
- ↑ http://news.mit.edu/2018/machines-learn-language-human-interaction-1031
- ↑ https://en.wikipedia.org/wiki/Natural_language_processing
- ↑ https://en.wikipedia.org/wiki/Statistical_inference
- ↑ https://en.wikipedia.org/wiki/Natural_language_processing
- ↑ https://en.wikipedia.org/wiki/Heuristic_(computer_science)