Machine Learning Seminar (5)
Download as PPTX, PDF0 likes142 views
Basics of Natural Language Processing
![箭猟
1. This movie is terrible. It¨s a waste of time.
2. This movie was good and made me happy. Had a very good time.
3. This movie is just boring.
1. [movie, terrible, waste, time]
2. [movie, good, made, happy, good, time]
3. [movie, boring]
18
念I尖](https://image.slidesharecdn.com/machine-learning-seminar-5-200619073543/85/Machine-Learning-Seminar-5-18-320.jpg)
![[恷仟井] JSAI2018 チュ`トリアル仝"侮嗔Яr旗の" ゼロから兵める徭隼冱囂I尖々](https://cdn.slidesharecdn.com/ss_thumbnails/jsai2018tutorialslideshare-180607030706-thumbnail.jpg?width=560&fit=bounds)
More Related Content
Similar to Machine Learning Seminar (5) (20)
Recently uploaded (11)
Machine Learning Seminar (5)
- 1. C亠僥茶氏(5) 徭隼冱囂I尖の児A June 18, 2020 Tomoya Nakayama
- 2. 書晩h苧すること ? 徭隼冱囂I尖の畠悶 ? テキストの念I尖 ? テキストデ`タの蒙翮燭臨り圭 2
- 4. 1. 畠悶の送れ 4
- 6. 徭隼冱囂I尖における畠悶の送れ テキストの念I尖 ? 音勣な勣殆の 茅肇 ? gZ蛍護 ? 屎ｻ ? ´など 蒙翮燭臨撹 ? 竃F指方 ? 竃Fl業(TF) ? TF-IDF ? word2vec ? BERT ? ´など モデルB ? 侘指｢ ? 屈ｷ幗 ? マルチクラス 蛍 ? ´など 6
- 7. ‐畽／デ`タの念I尖 ? ダミ`晒|議篳を0/1の方､Q ? 之p?翌れ､I尖 ? 屎ｻ､0゛1にQ ? 併辺ｷ峅爾惄錣擦襭 7
- 8. ‐畽／屈ｷ幗 親朕1 親朕2 栽倦 64 75 〜 77 66 ＄ 82 55 〜 49 95 ＄ 58 83 ＄ 69 69 ? 40 50 60 70 80 90 100 40 50 60 70 80 90 8 栽鯉と音栽鯉の 廠順をつける
- 9. ‐畽／屈ｷ幗の尖 ? 朕議v方 ? ? = ?(? ? ? + ?) ? σはシグモイド(sigmoid)v方 ? p払v方 ? 住餓エントロピ` ? ? = 1 ? ?=1 ? (??? log ?? ? 0 0.5 1 -6 -4 -2 0 2 4 6 y x シグモイドv方 9
- 10. 2. テキストの念I尖 10
- 11. 念I尖の麼な返隈 ? 音勣な勣殆の茅肇 ? gZ蛍護 ? gZの屎ｻ 11
- 12. 箭猟 12 The ship may have sunk but the movie didn't!!! Director, James Cameron, from 'The Terminator' did it again with this amazing picture. One of my favorite scenes is 'The Dinner table' scene, in which Rose's family and friends meet Jack after he saves her. Rose has a look on her face that every woman should have when you meet 'THE ONE'...I hope I have that look when I am in the room with my future husband.<br /><br />Jack and Rose have a connection that is 'MOVIE STUFF' but it's good movie stuff. We have the greedy mom and all her elite stuck up associates who live off of their husbands wealth. Rose almost commits suicide but the Gilbert Grape star rescues her. I really liked the hanging over the boat scene. It was a good risk.<br /><br />The movie is long but it's fantastic!!! Good story, good flow, good actors!!! Go see it twice if you want, Its worth it!!! 竃灸 Andrew L. Maas, Raymond E. Daly, Peter T. Pham, Dan Huang, Andrew Y. Ng, and Christopher Potts. (2011). Learning Word Vectors for Sentiment Analysis. The 49th Annual Meeting of the Association for Computational Linguistics (ACL 2011).
- 13. 音勣な勣殆の茅肇 ? 鞘i泣や催 ? ストップワ`ド聞喘l業が互すぎて蒙罿箸靴堂止mなもの ? 哂Zでは酵~ (a, the)?旗兆~ (I, you, they, ´)?念崔~ (in, on, of, ´)など ? 猟嫗とはvSない秤鵤HTMLタグなど ? 方忖 13
- 15. 屎ｻ (normalization) ? 寄猟忖を弌猟忖にQ ? 玉s侘の婢_ ? didn¨t -> did not ? Zﾖ晒(stemming) ´俊^看?俊硫看を函り茅く ? scenes -> scene, played -> play ? レンマ晒(lemmatization) ´Zの圻侘にQする ? sunk -> sink, knives -> knife 15
- 16. 3. 蒙翮燭臨撹 16
- 17. 蒙翮迅撹の麼な返隈 ? テキストのベクトル晒 ? bag-of-words ? TF ? IF-IDF ? gZのベクトル晒 ? word2vec ? GloVe ? BERT ? テキストとgZのベクトル晒 ? doc2vec┰饂悗牢イ譴覆ぃ 17
- 18. 箭猟 1. This movie is terrible. It¨s a waste of time. 2. This movie was good and made me happy. Had a very good time. 3. This movie is just boring. 1. [movie, terrible, waste, time] 2. [movie, good, made, happy, good, time] 3. [movie, boring] 18 念I尖
- 19. 竃F指方 (bag of words) ? テキストごとの光gZの竃F指方をそのまま蒙翮燭箸垢 19 boring good happy made movie terrible time waste Text 1 0 0 0 0 1 1 1 1 Text 2 0 2 1 1 1 0 1 0 Text 3 1 0 0 0 1 0 0 0
- 20. 竃Fl業(TF: Term Frequency) ? テキストごとの光gZの竃F指方を畠gZ方で護った､ 蒙翮燭箸垢 ? ???,? = テキスト ?におけるgZ ?の竃F指方 テキスト ?の畠gZ方 20 boring good happy made movie terrible time waste Text 1 0.00 0.00 0.00 0.00 0.25 0.25 0.25 0.25 Text 2 0.00 0.33 0.17 0.17 0.17 0.00 0.17 0.00 Text 3 0.50 0.00 0.00 0.00 0.50 0.00 0.00 0.00
- 21. TF-IDF (IDF: Inverse Document Frequency) ? TFにし、ほぼすべてのテキストに竃Fするような 仝レア業の詰い々gZの嶷みを和げる ? ???? = log 畠テキスト方 gZ ?が竃Fするテキスト方 ? ??????,? = ???,? 〜 ???? 21 boring good happy made movie terrible time waste IDF 0.48 0.48 0.48 0.48 0.00 0.48 0.30 0.48 Text 1 0.00 0.00 0.00 0.00 0.00 0.12 0.08 0.12 Text 2 0.00 0.16 0.08 0.08 0.00 0.00 0.05 0.00 Text 3 0.24 0.00 0.00 0.00 0.00 0.00 0.00 0.00
- 22. gZのベクトル晒 ? gZを詰肝圷のベクトルで燕F ? このベクトルを仝gZ托めzみ(word embedding)々とも冱う ? 繁議な湖では50゛200肝圷くらい ? 仝除い々gZはベクトルが除くなる ? gZのベクトルをMみ栽わせてテキストのベクトルを恬る ? gZのベクトルの栽or峠譲をとる ? gZのベクトルをBYさせる畠デ`タのLさを栽わせる駅勣あり ? 僥gみのベクトルが巷_されていることも 22
- 23. word2vec ? Distributed Representation of Words and Phrases and their Compositionality (Mikolov et al., 2013) ? 慌軟する┰くにFれるg Zを僥させる ? ベクトルどうしのvS來もベ クトルで燕せる ? Paris C France + Japan = ? 23
- 24. その麿のgZ托めzみモデル ? GloVe: Global Vectors for Word Representation ? Pennington et al., 2014 ? BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding ? Devlin, et al. 2018 24
- 25. 4. ハンズオン 25
- 26. ハンズオン ? Google Colab井 ? ハンズオンのg佩にはGoogle アカウントが駅勣です。 ? 書指からBinder井は峭しま した^肇に旋喘宀がいな かったため 26