일 | 월 | 화 | 수 | 목 | 금 | 토 |
---|---|---|---|---|---|---|
1 | 2 | 3 | ||||
4 | 5 | 6 | 7 | 8 | 9 | 10 |
11 | 12 | 13 | 14 | 15 | 16 | 17 |
18 | 19 | 20 | 21 | 22 | 23 | 24 |
25 | 26 | 27 | 28 | 29 | 30 | 31 |
- nlp
- Python
- 자기조직화지도
- TFX
- ResNet
- stemming
- Binary classification
- RNN
- Ann
- 경사하강법
- tensorflow
- cross domain
- NER
- SOMs
- LSTM
- Logistic Regression
- Transfer Learning
- 군집화
- NMF
- textmining
- Attention
- BERT
- gaze estimation
- VGGNet
- Support Vector Machine
- Generative model
- MLOps
- AI 윤리
- Clustering
- Gradient Descent
- Today
- Total
목록Clustering (3)
juooo1117

Clustering with NMF(Non-negative factorization) Document Classification 참고 블로그 : https://www.davidsbatista.net/blog/2017/04/01/document_classification/ - classifiy a document into a set of pre-defined categories using different supervised classifiers and text representations. - Only use the plot of the movie and the genre on which the movie is classified. - Dataset: IMDB datasets of movie, genre..

Clustering data instances를 비슷한 것끼리 묶어서 그룹을 만드는 방법으로 같은 cluster에 속해있는 문서들이라면 해당 문서들은 서로 비슷해야 하며, 다른 cluster에 해당되는 문서들끼리는 서로 달라야 한다. unsupervised learning의 한 방법이다. Non-negative Matrix Factorization(NMF) 비음수 행렬 분해: 행렬을 분해하는 방법으로 모든 요소가 음수가 아닌 행렬 V를, 모든 요소가 음수가 아닌 행렬 W, H의 곱으로 분해한다. NMF는 document clustering 에 사용할 수 있다. Goal: V가 주어졌을 때, V ≈ WH 를 만족하는 W, H 를 구해내는 것 V의 column vector는 W의 column vector들과..
Clustering데이터 instance들을 비슷한 것끼리 묶어서 그룹을 만드는 기법데이터가 문서일 때는 문서 클러스터링같은 cluster에 있는 문서들은 서로 비슷하고, 다른 cluster에 있는 문서들은 서로 달라야함비지도학습(Unsupervised Learning의 한 방법 → 즉, label이 없는 상태) Similarity문서 clustering에서 비슷한 문서끼리 묶는다고 했을 때, 비슷함은 무엇을 기준으로 잡을까?→ vetor space에서 가까운 거리에 있는 것들! (Euclidean Distance)→ 확률적으로 비슷한 단어들이 많이 발생하는 문서들 K-means clustering간단하면서도 성능이 좋은 대표적인 clustering 방법론으로, 각 cluster는 cluster cent..