Classifying Perceived Emotions based on Polarity of Arousal and Valence from Sound Events. Krishan, P. & Abri, F. In 2022 IEEE International Conference on Big Data (Big Data), pages 2849–2856, December, 2022. doi abstract bibtex Sonification uses sounds to glean insights about information and activities in a person’s life. There are two types of emotions based on sounds: perceived emotions and induced emotions. This paper focuses on classifying perceived emotions based on two dimensions – arousal and valence, using several deep-learning models. Four feature selection techniques, Forward Feature Selection, Recursive Feature Elimination, Random Forest, and Principal Component Analysis, are performed; class imbalance in the dataset is demonstrated and handled using under-sampling, and over-sampling techniques, and the results are compared. This paper shows the need for balanced data to train classifiers and the advantages of running classifiers on the balanced dataset that is generated using sampling techniques. The eXtreme Gradient Boosting (XgB) classifier trained and tested on the over-sampled balanced dataset using all the features generates a test F1 score of 81.5 and is the best model that can be selected from all the classifiers.
@inproceedings{krishan_classifying_2022,
title = {Classifying {Perceived} {Emotions} based on {Polarity} of {Arousal} and {Valence} from {Sound} {Events}},
doi = {10.1109/BigData55660.2022.10020353},
abstract = {Sonification uses sounds to glean insights about information and activities in a person’s life. There are two types of emotions based on sounds: perceived emotions and induced emotions. This paper focuses on classifying perceived emotions based on two dimensions – arousal and valence, using several deep-learning models. Four feature selection techniques, Forward Feature Selection, Recursive Feature Elimination, Random Forest, and Principal Component Analysis, are performed; class imbalance in the dataset is demonstrated and handled using under-sampling, and over-sampling techniques, and the results are compared. This paper shows the need for balanced data to train classifiers and the advantages of running classifiers on the balanced dataset that is generated using sampling techniques. The eXtreme Gradient Boosting (XgB) classifier trained and tested on the over-sampled balanced dataset using all the features generates a test F1 score of 81.5 and is the best model that can be selected from all the classifiers.},
booktitle = {2022 {IEEE} {International} {Conference} on {Big} {Data} ({Big} {Data})},
author = {Krishan, Pooja and Abri, Faranak},
month = dec,
year = {2022},
keywords = {Big Data, Boosting, Classification, Cyber-physical systems, Emotion prediction, Feature extraction, Perceived emotions, Principal component analysis, Sonification, Sound},
pages = {2849--2856},
}
Downloads: 0
{"_id":"ZhicWhz79iPkWpzs7","bibbaseid":"krishan-abri-classifyingperceivedemotionsbasedonpolarityofarousalandvalencefromsoundevents-2022","author_short":["Krishan, P.","Abri, F."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Classifying Perceived Emotions based on Polarity of Arousal and Valence from Sound Events","doi":"10.1109/BigData55660.2022.10020353","abstract":"Sonification uses sounds to glean insights about information and activities in a person’s life. There are two types of emotions based on sounds: perceived emotions and induced emotions. This paper focuses on classifying perceived emotions based on two dimensions – arousal and valence, using several deep-learning models. Four feature selection techniques, Forward Feature Selection, Recursive Feature Elimination, Random Forest, and Principal Component Analysis, are performed; class imbalance in the dataset is demonstrated and handled using under-sampling, and over-sampling techniques, and the results are compared. This paper shows the need for balanced data to train classifiers and the advantages of running classifiers on the balanced dataset that is generated using sampling techniques. The eXtreme Gradient Boosting (XgB) classifier trained and tested on the over-sampled balanced dataset using all the features generates a test F1 score of 81.5 and is the best model that can be selected from all the classifiers.","booktitle":"2022 IEEE International Conference on Big Data (Big Data)","author":[{"propositions":[],"lastnames":["Krishan"],"firstnames":["Pooja"],"suffixes":[]},{"propositions":[],"lastnames":["Abri"],"firstnames":["Faranak"],"suffixes":[]}],"month":"December","year":"2022","keywords":"Big Data, Boosting, Classification, Cyber-physical systems, Emotion prediction, Feature extraction, Perceived emotions, Principal component analysis, Sonification, Sound","pages":"2849–2856","bibtex":"@inproceedings{krishan_classifying_2022,\n\ttitle = {Classifying {Perceived} {Emotions} based on {Polarity} of {Arousal} and {Valence} from {Sound} {Events}},\n\tdoi = {10.1109/BigData55660.2022.10020353},\n\tabstract = {Sonification uses sounds to glean insights about information and activities in a person’s life. There are two types of emotions based on sounds: perceived emotions and induced emotions. This paper focuses on classifying perceived emotions based on two dimensions – arousal and valence, using several deep-learning models. Four feature selection techniques, Forward Feature Selection, Recursive Feature Elimination, Random Forest, and Principal Component Analysis, are performed; class imbalance in the dataset is demonstrated and handled using under-sampling, and over-sampling techniques, and the results are compared. This paper shows the need for balanced data to train classifiers and the advantages of running classifiers on the balanced dataset that is generated using sampling techniques. The eXtreme Gradient Boosting (XgB) classifier trained and tested on the over-sampled balanced dataset using all the features generates a test F1 score of 81.5 and is the best model that can be selected from all the classifiers.},\n\tbooktitle = {2022 {IEEE} {International} {Conference} on {Big} {Data} ({Big} {Data})},\n\tauthor = {Krishan, Pooja and Abri, Faranak},\n\tmonth = dec,\n\tyear = {2022},\n\tkeywords = {Big Data, Boosting, Classification, Cyber-physical systems, Emotion prediction, Feature extraction, Perceived emotions, Principal component analysis, Sonification, Sound},\n\tpages = {2849--2856},\n}\n\n","author_short":["Krishan, P.","Abri, F."],"key":"krishan_classifying_2022","id":"krishan_classifying_2022","bibbaseid":"krishan-abri-classifyingperceivedemotionsbasedonpolarityofarousalandvalencefromsoundevents-2022","role":"author","urls":{},"keyword":["Big Data","Boosting","Classification","Cyber-physical systems","Emotion prediction","Feature extraction","Perceived emotions","Principal component analysis","Sonification","Sound"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/fsimonetta","dataSources":["pzyFFGWvxG2bs63zP"],"keywords":["big data","boosting","classification","cyber-physical systems","emotion prediction","feature extraction","perceived emotions","principal component analysis","sonification","sound"],"search_terms":["classifying","perceived","emotions","based","polarity","arousal","valence","sound","events","krishan","abri"],"title":"Classifying Perceived Emotions based on Polarity of Arousal and Valence from Sound Events","year":2022}