SSNaturalLanguage

The SSNaturalLanguage framework let you provide a variety of natural language processing functionality with support for many different languages and scripts. Use SSNaturalLanguage framework to segment natural language text into paragraphs, sentences or words and tag information about those segments such as part of speech, lexical class, lemma, script and language.

Requirements

• iOS 13.0+
• Xcode 11+

Installation

CocoaPods

• You can use CocoaPods to install SSNaturalLanguage by adding it to your Podfile:

   use_frameworks!
   pod 'SSNaturalLanguage'
  

• import SSNaturalLanguage

Manually

• Download and drop SSNaturalLanguage folder in your project.
• Congratulations!

Swift Package Manager

• When using Xcode 11 or later, you can install SSNaturalLanguage by going to your Project settings > Swift Packages and add the repository by providing the GitHub URL. Alternatively, you can go to File > Swift Packages > Add Package Dependencies...

dependencies: [
    .package(url: "https://github.com/SimformSolutionsPvtLtd/SSNaturalLanguage.git", from: "1.0.0")
]

Carthage

• Carthage is a decentralized dependency manager that builds your dependencies and provides you with binary frameworks. You can install Carthage with Homebrew using the following command:

$ brew update
$ brew install carthage

To integrate SSNaturalLanguage into your Xcode project using Carthage, add the following line to your Cartfile:

github "SimformSolutionsPvtLtd/SSNaturalLanguage"

Run carthage to build and drag the SSNaturalLanguage(Sources/SSNaturalLanguage) into your Xcode project.

How It Works

• SSNaturalLanguage framework let you help with language related stuffs like tokenisation, lemmatization, languageIdentification, spellCorrection, partOfSpeech, sentimental score, neighboringWord, uniqueTagsFromSentense, etc... with just a single function call.

1. Split Text

• SSNaturalLanguage can split a string into chunks of words, sentences, characters or paragraphs. These segmented texts can then be processed together or separately, depending on the use case. To split string, the NLTokenizer class is used.

import SSNaturalLanguage
var text = "How are you? Where were you?"
let arrSplitText = text.splitInto(unit: .sentence)

• splitInto function will return array of unit passed in input param.

2. Base Form

• SSNaturalLanguage can convert a word into its base form. For example "assumed" and "assuming" will convert into "assume".
• It is commonly used in word tagging and fuzzy matching (which identifies misspelled words, like what you'd see in a Google search).

var text = "I am running late for the 10 km marathon run that was scheduled for today. Can we reschedule the meeting?"
let arrBaseForm = text.toBaseForm()

• It will return array of tuple with word and its baseForm.

I - I
am - be
running - run
late - late
for - for
the - the
marathon - marathon
run - run
that - that
was - be
scheduled - schedule
for - for
today - today
Can - can
we - we
reschedule - reschedule
the - the
meeting - meeting

3. Language Identification

• identifyLanguage function helps us determine a piece of text language from a string. It returns language code from the text's context.

var text = "Gracias"
let languageCode = text.identifyLanguage()
print(languageCode) ==> "es"

• identifyLanguage will return language code. we can now convert it to particular language name with:

Locale.current.localizedString(forIdentifier: languageCode)) ==> "Spanish"

4. Predicted Language

• Following predictedLanguage function returns the top 2 predicted languages and return array of tuple with languageCode and its confidence.

var text = "Mart Goorsqu'ent avainq ce mennoci faucte. pirrin to l'onçair-là der mosepe, qués, audeu! of therci; this asiblio whot fore; Mme jecome de da vidger. Cohe witid-hus joir."
let arrPredictedLangs = text.predictedLanguage(maxPredictCount: 2)

5. Spell Correction

• Spell Checking and correction is another very important and popular application that has real-world value for any text-based apps that you might build.
• The biggest example of this is Google search itself. it tries to correct our spelling to make sure we get the right results for your query.
• correctSpell function will help you to correct spelling from the text.

let text = """
I started my schooling as the majority did in my area, at the local primarry school. I then
went to the local secondarry school and recieved grades in English, Maths, Phisics,
Biology, Geography, Art, Graphical Comunication and Philosophy of Religeon. I'll not
bore you with the 'A' levels and above.
"""
let correctText = text.correctSpell()

Result: I started my schooling as the majority did in my area, at the local primary school. I then went to the local secondary school and received grades in English, Maths, Physics, Biology, Geography, Art, Graphical Communication and Philosophy of Religion. I'll not bore you with the 'A' levels and above.

6. Part Of Speech (POS) Tagging

• Every word in a sentence is associated with a Part Of Speech (POS) tag - noun, verbs, adjectives, adverbs, etc.
• POS tags are widely used for text analytics as they encompass additional information of the text data at hand.
• partOfSpeech function will diagnosis input text and return array of tuple with word and its tag.

let text = "Hello world, I am a data scientist. I work with machine learning!"
let pos = text.partOfSpeech()

Result: 
Hello - Interjection
world - Noun
I - Pronoun
am - Verb
a - Determiner
data - Noun
scientist - Noun
I - Pronoun
work - Verb
with - Preposition
machine - Noun
learning - Noun

7. Identify person Entity

• Following recognizePersonName will return array of identified personName from given text.

let text = "Jackie, are you leaving so soon?"
let arrPersonName = text.recognizePersonName()

Result:
["Jackie"]

8. Identify placeName

• Following recognizePlaceName will return array of identified placeName from given text.

let text = "Apple is looking at buying U.K. startup for $1 billion."
let arrPlaceName = text.recognizePlaceName()

Result:
["U.K."]

9. Identify Organization Name

• Following recognizeOrganizationName will return array of identified organizationName from given text.

let text = "Apple is looking at buying U.K. startup for $1 billion."
let arrOrganizationName = text.recognizeOrganizationName()

Result:
["Apple"]

10. Sentimental Analysis

• Sentiment Analysis is when we try to predict the sentiment of a given piece of text: is it positive, negative or neutral?
• sentimentalScore function will return sentimental score from given text.
• The range of sentimental score is [-1.0, 1.0]. A score of 1.0 is the most positive, a score of -1.0 is the most negative, and a score of 0.0 is neutral.
• Sentiment analysis feature is supports 7 languages: English, Spanish, French, Italian, German, Portuguese, Simplified Chinese.

let text = "I hate this apple pie."
let sentimentalScore = text.sentimentalScore()
print(sentimentalScore) ==> -0.6

11. Word Embedding

• neighboringWords function will find similar kinds of word from given word. Ex. Let's say we type "king". Off the top of my head, we would want words like "prince", "crown", "throne" etc...

let word = "cheese"
let arrSimilarWords = word.neighboringWords(maximumResult: 5)

Result:
[(word: "mozzarella", distance: 0.631902277469635), (word: "cheddar", distance: 0.6747748851776123), (word: "provolone", distance: 0.6827872395515442), (word: "ricotta", distance: 0.6940725445747375), (word: "focaccia", distance: 0.7102253437042236)]

11. Unique Tags

• To find unique word from given string, findUniqueTag will help.

let text = "I love machine learning and I work as a Data Scientist in India."
let arrUniqueWord = text.findUniqueTag()

Result:
["Machine", "Learning", "Scientist", "India", "Data"]

Find this library useful? ❤️

Support it by joining stargazers for this repository. ⭐

License

• SSNaturalLanguage is available under the MIT license. See the LICENSE file for more info.