Exploring Natural Language Processing
summary
Demo shows iOS 11's Core ML plus Foundation’s NSLinguisticTagger doing tokenization, language identification, part‑of‑speech tagging, lemmatization for search, and named‑entity recognition; a sample app and playground on GitHub extract verbs, nouns, joined multiword names, and match plural/tense variants for search. It runs fast and on-device privacy is preserved; model updates arrive via keyboard downloads. Limitations include limited NER language coverage (fine‑grain tags in ~8 languages) and misses like Twitter/Facebook/BuzzFeed, trouble with mixed‑language sentences, and brittle single‑word tagging.
AI-written summary · about
chapters (25)
- 0:00Intro
- 0:50What is Natural Language Processing
- 1:13Core ML
- 2:12NLP API
- 2:32Who should use the NLP API
- 2:57What is useful information
- 3:23Tokenization
- 3:58Limitization
- 4:26Entity Recognition
- 5:15Linguistic Tags
- 5:59Type Tags
- 6:27Chart
- 6:52Language Identification
- 7:34Counting Tokens
- 9:50Sample App
- 10:27Named Entity
- 11:04Limit ization
- 12:22Sample project
- 12:39Benefits of NLP API
- 12:55Performance
- 13:35What didnt work
- 13:45Named Entity Identification
- 14:20Spanglish
- 15:34Examples
- 16:36Outro
related talks
31mPower-Up Your App With Apple’s Machine Learning Frameworks
34mVoice Search Using NLP & CoreML
43mGetting Sentimental: exploration of iOS sentiment analysis
55mDemystifying Natural Language Processing across several industry verticals
24mParsing natural languages in Swift
01h 09mAutomatic Tagging of Short Texts with scikit-learn and NLTK
55mKeynote - Natural Language Processing: Challenges and Next Frontiers
39mCustomize Your App with MLKit