Emotion Detection in Text using Nested Long Short-Term Memory
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2019, Vol 10, Issue 6
Abstract
Humans have the power to feel different types of emotions because human life is filled with many emotions. Human’s emotion can be reflected through reading or writing a text. In recent years, studies on emotion detection through text has been developed. Most of the study is using a machine learning technique. In this paper, we classified 7 emotions such as anger, fear, joy, love, sadness, surprise, and thankfulness using deep learning technique that is Long Short-Term Memory (LSTM) and Nested Long Short-Term Memory (Nested LSTM). We have compared our results with Support Vector Machine (SVM). We have trained each model with 980,549 training data and tested with 144,160 testing data. Our experiments showed that Nested LSTM and LSTM give better performance than SVM to detect emotions in text. Nested LSTM gets the best accuracy of 99.167%, while LSTM gets the best performance in term of average precision at 99.22%, average recall at 98.86%, and f1-score at 99.04%.
Authors and Affiliations
Daniel Haryadi, Gede Putra Kusuma
OSPF vs EIGRP: A Comparative Analysis of CPU Utilization using OPNET
Routing is difficult in enterprise networks because a packet might have to traverse many intermediary nodes to reach the final destination. The selection of an appropriate routing protocol for a large network is difficul...
Knowledge discovery from database using an integration of clustering and classification
Clustering and classification are two important techniques of data mining. Classification is a supervised learning problem of assigning an object to one of several pre-defined categories based upon the attributes of the...
Double Diode Ideality Factor Determination using the Fixed-Point Method
In this paper, we are interested in the diode ideality factor study of the double exponential equivalent model, based on the properties of the fixed point method. The optimal choice of this factor will improve the photov...
Designing Graphical Data Storage Model for Gene-Protein and Gene-Gene Interaction Networks
Graph is an expressive way to represent dynamic and complex relationships in highly connected data. In today’s highly connected world, general purpose graph databases are providing opportunities to experience benefits of...
Development of Self-Learning Program for the Bending Process of Quartz Glass
Quartz glass is a high-performance glass material with its high heat and chemical resistance, wide optical transparency ranging from ultraviolet light to infrared light, and the high formativeness as a glass material. Be...