Detecting self-esteem level and depressive indication due to different parenting style using supervised learning techniques

dc.contributor.advisorAlam, Md. Golam Rabiul
dc.contributor.advisorRoy, Shaily
dc.contributor.authorAl Taawab, Abdullah
dc.contributor.authorRahman, Mahfuzzur
dc.contributor.authorIslam, Zawadul
dc.contributor.authorMustari, Nafisa
dc.date.accessioned2022-09-14T05:21:26Z
dc.date.available2022-09-14T05:21:26Z
dc.date.issued2022-05
dc.descriptionCataloged from PDF version of thesis.
dc.descriptionIncludes bibliographical references (pages 37-39).
dc.descriptionThis thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2022.
dc.description.abstractUprising a child is a psychological construct of parents, which is a combination of factors that evolves over time with the growth and development of the child. Parent ing style represents a set of strategies that have diverse influences on children. These approaches can create depressive symptoms in children’s minds, which can last even if they become adolescents. Moreover, these indications may affect their level of self confidence. In this research, supervised learning models are used to detect different parenting styles, depression indications of adolescents due to parenting and the level of their self-esteem. Due to the absence of publicly available data, we created our own data set of about 500 survey responses. Additionally, eleven psychological and nine linguistic attributes of Linguistic Inquiry and Word Count (LIWC) have been used to identify depression indications. Among all the supervised models, the Lo gistic Regression (LR), Gradient Boost Classifier (GBC) and Bi-Directional LSTM (Bi-LSTM) provide better results than other models. This research is capable of helping the parents to know their children’s psychology in a better way and make them have a more profound discussion on practical life.
dc.identifier.otherID: 20341043
dc.identifier.otherID: 18101485
dc.identifier.otherID: 18101005
dc.identifier.otherID: 19101086
dc.identifier.otherhttps://dspace.bracu.ac.bd/server/api/core/items/3ffd44c1-7281-46bd-bd34-68033366dd00
dc.identifier.urihttp://hdl.handle.net/10361/17215
dc.language.isoen_US
dc.publisherBRAC University
dc.sourceBRAC University Institutional Repository
dc.subjectMachine Learning
dc.subjectDeep Learning
dc.subjectLIWC
dc.subjectNLP
dc.subjectDepression
dc.subjectParenting style
dc.subjectSelf Esteem
dc.titleDetecting self-esteem level and depressive indication due to different parenting style using supervised learning techniques
dc.typeThesis

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
20341043, 18101485, 18101005, 19101086_CSE.pdf
Size:
954.18 KB
Format:
Adobe Portable Document Format