Seminar presentation made by me for the topic of 'Resources for Sentiment Analysis' at IIT Bombay. Includes a set of bonus slides for additional information which was not actually presented.
Following are some suggestions for future research. As GFRSCC technology is now being adopted in many countries throughout the world, in the absence of suitable
standardized test methods it is necessary to examine the existing test methods and identify or, when necessary, develop test methods suitable for acceptance as International Standards. Such test methods have to be capable of a rapid and reliable assessment of key
properties of fresh SCC on a construction site. At the same time, the testing equipment should be reliable, easily portable and inexpensive. The test procedure should be carried out by a single operator and the test results have to be interpreted with a minimum of training. Also, the results have to define and specify different GFRSCC mixes. One primary application of these test methods would be in verification of compliance on sites and in concrete production plants, if self- compacting concrete could be manufactured in large quantities..
Seminar presentation made by me for the topic of 'Resources for Sentiment Analysis' at IIT Bombay. Includes a set of bonus slides for additional information which was not actually presented.
Following are some suggestions for future research. As GFRSCC technology is now being adopted in many countries throughout the world, in the absence of suitable
standardized test methods it is necessary to examine the existing test methods and identify or, when necessary, develop test methods suitable for acceptance as International Standards. Such test methods have to be capable of a rapid and reliable assessment of key
properties of fresh SCC on a construction site. At the same time, the testing equipment should be reliable, easily portable and inexpensive. The test procedure should be carried out by a single operator and the test results have to be interpreted with a minimum of training. Also, the results have to define and specify different GFRSCC mixes. One primary application of these test methods would be in verification of compliance on sites and in concrete production plants, if self- compacting concrete could be manufactured in large quantities..
Robust Pathway-based Multi-Omics Data Integration using Directed Random Walk ...SOYEON KIM
17th Annual International Conference on Critical Assessment of Massive Data Analysis (CAMDA 2018)
Cancer Data Integration Challenge (http://camda.info/)
Deep learning based multi-omics integration, a surveySOYEON KIM
1. Unsupervised feature construction and knowledge extraction from genome-wide assays of breast cancer with denoising autoencoders, Pacific Symposium on Biocomputing, 2015
2. A deep learning approach for cancer detection and relevant gene identification, Pacific Symposium on Biocomputing, 2016
3. Deep Learning based multi-omics integrationrobustly predicts survival in liver cancer, preprint, 2017
Pathways-Driven Sparse Regression Identifies Pathways and Genes Associated wi...SOYEON KIM
Summary of paper "Pathways-Driven Sparse Regression Identifies Pathways and Genes Associated with High-Density Lipoprotein Cholesterol in Two Asian Cohorts",
Silver M, Chen P, Li R, Cheng C-Y, Wong T-Y, et al.
In PLOS Genetics, 2013
A survey of heterogeneous information network analysisSOYEON KIM
A Survey of Heterogeneous Information Network Analysis
Chuan Shi, Member, IEEE,
Yitong Li, Jiawei Zhang, Yizhou Sun, Member, IEEE,
and Philip S. Yu, Fellow, IEEE
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2015
Translated Learning: Transfer learning across different feature spaces
Wenyuan Dai, Yuqiang Chen, Gui-Rong Xue, Qiang Yang, and Yong Yu.
In Proceedings of Twenty-Second Annual Conference on Neural Information Processing Systems (NIPS 2008)
Semi-automatic ground truth generation using unsupervised clustering and limi...SOYEON KIM
Semi-automatic ground truth generation using unsupervised clustering and limited manual labeling: Application to handwritten character recognition
Szilárd Vajda, Yves Rangoni, Hubert Cecotti
Pattern Recognition Letters, 2015
Evaluating color descriptors for object and scene recognitionSOYEON KIM
Van De Sande, Koen EA, Theo Gevers, and Cees GM Snoek. "Evaluating color descriptors for object and scene recognition." Pattern Analysis and Machine Intelligence, IEEE Transactions on 32.9 (2010): 1582-1596.
Outcome-guided mutual information networks for investigating gene-gene intera...SOYEON KIM
TBC2014 poster
"Outcome-guided mutual information networks for investigating gene-gene interaction effects on clinical outcomes", Hyun-hwan Jeong, So Yeon Kim, Kyubum Wee, Kyung-Ah Sohn
Robust Pathway-based Multi-Omics Data Integration using Directed Random Walk ...SOYEON KIM
17th Annual International Conference on Critical Assessment of Massive Data Analysis (CAMDA 2018)
Cancer Data Integration Challenge (http://camda.info/)
Deep learning based multi-omics integration, a surveySOYEON KIM
1. Unsupervised feature construction and knowledge extraction from genome-wide assays of breast cancer with denoising autoencoders, Pacific Symposium on Biocomputing, 2015
2. A deep learning approach for cancer detection and relevant gene identification, Pacific Symposium on Biocomputing, 2016
3. Deep Learning based multi-omics integrationrobustly predicts survival in liver cancer, preprint, 2017
Pathways-Driven Sparse Regression Identifies Pathways and Genes Associated wi...SOYEON KIM
Summary of paper "Pathways-Driven Sparse Regression Identifies Pathways and Genes Associated with High-Density Lipoprotein Cholesterol in Two Asian Cohorts",
Silver M, Chen P, Li R, Cheng C-Y, Wong T-Y, et al.
In PLOS Genetics, 2013
A survey of heterogeneous information network analysisSOYEON KIM
A Survey of Heterogeneous Information Network Analysis
Chuan Shi, Member, IEEE,
Yitong Li, Jiawei Zhang, Yizhou Sun, Member, IEEE,
and Philip S. Yu, Fellow, IEEE
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2015
Translated Learning: Transfer learning across different feature spaces
Wenyuan Dai, Yuqiang Chen, Gui-Rong Xue, Qiang Yang, and Yong Yu.
In Proceedings of Twenty-Second Annual Conference on Neural Information Processing Systems (NIPS 2008)
Semi-automatic ground truth generation using unsupervised clustering and limi...SOYEON KIM
Semi-automatic ground truth generation using unsupervised clustering and limited manual labeling: Application to handwritten character recognition
Szilárd Vajda, Yves Rangoni, Hubert Cecotti
Pattern Recognition Letters, 2015
Evaluating color descriptors for object and scene recognitionSOYEON KIM
Van De Sande, Koen EA, Theo Gevers, and Cees GM Snoek. "Evaluating color descriptors for object and scene recognition." Pattern Analysis and Machine Intelligence, IEEE Transactions on 32.9 (2010): 1582-1596.
Outcome-guided mutual information networks for investigating gene-gene intera...SOYEON KIM
TBC2014 poster
"Outcome-guided mutual information networks for investigating gene-gene interaction effects on clinical outcomes", Hyun-hwan Jeong, So Yeon Kim, Kyubum Wee, Kyung-Ah Sohn
2. 1. text의 주관성, 객관성 판단
-> SO polarity (Pang and Lee,
2004; Yu and Hatzivassiloglou,
2003)
2. 주관성을 지닌 text의 긍정, 부
정 판단
-> PN polarity (Pang and Lee,
2004; Turney, 2002)
3. 얼마나 긍정/부정인지 판단
예) 조금 긍정, 약간 긍정, 아주
긍정
-> strength of text PN polarity
(Pang and Lee, 2005; Wilson et
al., 2004)
3. Polarity classification
Bo Pang and Lillian Lee, A Sentimental Education: Sentiment Analysis
Using Subjectivity Summarization Based on Minimum Cuts
4. WordNet
synset
- 영어의 의미 어휘
목록
- synset (유의어 집
단)으로 분류하여
단어집과 유의어,
반의어 사전의 배
합을 만듬.
- 심리학 교수인 조
지 A. 밀러가 지도
하는 프린스턴 대
학의 인지 과학 연
구소에 의해 만들
어졌고, 유지되고
있음.
9. Lp, Ln, Lo -> vectorial representation -> label Ci
-> 2개의 classifier 생성
1. Positive / not Positive 로 분류하는 classifier
2. Negative / not Negative 로 분류하는 classifier
Classifier
“Supervised learner”
Positive
∩
10. Lp, Ln, Lo -> vectorial representation -> label Ci
-> 2개의 classifier 생성
1. Positive / not Positive 로 분류하는 classifier
2. Negative / not Negative 로 분류하는 classifier
Classifier
“Supervised learner”
Negative
∩
11. Lp, Ln, Lo -> vectorial representation -> label Ci
-> 2개의 classifier 생성
1. Positive / not Positive 로 분류하는 classifier
2. Negative / not Negative 로 분류하는 classifier
Classifier
“Supervised learner”
Objective
∩ ∩
∪
12. Precision = Tp / (Tp + Fp)
: True라고 예측한 것 중에서 실제로 true인 것의 비율
Recall = Tp / (Tp + Fn)
: 실제로 true인 것중에 내가 얼마나 맞췄는지
(Tp : true라고 예측했는데 실제로 true,
Fp : true라고 예측했는데 실제로 false,
Fn : false라고 예측했는데 실제로 true,
Tn : false라고 예측했는데 실제로 false)
K를 정하려면?
K
Precision
Recall
Small
Training set
noise
(Andrea Esuli1 and Fabrizio Sebastiani2, Determining Term Subjectivity and
Term Orientation for Opinion Mining)