Classification of Short-Texts Generated During Disasters: Traditional and Deep learning Approach

Kundu, Shamik and Srijith, P K (2018) Classification of Short-Texts Generated During Disasters: Traditional and Deep learning Approach. Masters thesis, Indian Institute of Technology Hyderabad.

Thesis_Mtech_CS_4096.pdf - Published Version

Download (1MB) | Preview


Micro-blogging sites provide a wealth of resources during disaster events in the form of short texts. Correct classification of those short texts into various actionable classes can be of great help in shaping the means to rescue people in disaster-a�ected places. The process of classification of short texts poses a challenging problem because the texts are usually short and very noisy and Inding good features that can distinguish these texts into di�erent classes is time consuming, tedious and often requires a lot of domain knowledge. In this thesis, we explore various non-deep learning and deep learning methods and propose a deep learning based model to classify tweets into difierent actionable classes such as resource need and availability, activities of various NGO etc. The proposed model requires no domain knowledge and can be used in any disaster scenario with little to no modification. Keywords: Text classification, Topic Modelling, LDA, Word-embeddings, LSTM, Deep Learning

[error in script]
IITH Creators:
IITH CreatorsORCiD
Item Type: Thesis (Masters)
Uncontrolled Keywords: Text Classification, Topic Modelling, LDA, Word embeddings
Subjects: Computer science
Divisions: Department of Computer Science & Engineering
Depositing User: Team Library
Date Deposited: 28 Jun 2018 09:47
Last Modified: 28 Jun 2018 09:47
Publisher URL:
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 4096 Statistics for this ePrint Item