Klasifikasi Topik dan Analisa Sentimen Terhadap Kuesioner Umpan Balik Universitas Menggunakan Metode Long Short-Term Memory

Doenny Auddy Lionovan, Leo Willyanto Santoso, Rolly Intan

Abstract


In evaluating and improving the quality of services and facilities, Petra Christian University (PCU) took feedback questionnaires given to students via online. At this time, reading the comments in the suggestion column is still read manually. So that it becomes less effective and efficient in analyzing sentiments and classifying topics for many comments. This study will apply word2vec and Long Short-Term Memory method to create a program that can help classify topics and sentiment analysis.

Word2vec is used as a method to convert a word into a vector along with mapping the meaning of existing words. The parameter tested on word2vec are the number of iterations and windows size. Whereas the Long Short-Term Memory method is used to classify sentiment and topics. The parameter tested are the number of layers, the number of units, the batch size, and the percentage of dropout.

The result of this study indicates that the word2vec method along with Long Short-Term Memory can be used to analyze sentiment and topic classification. The best configuration results obtained average accuracy in the implementation of the sentiment classification is 89,16 % and for implementation of the topic classification is 92,98 %.


Keywords


Natural Language Processing; Sentiment Analysis; Text Classification; Long Short-Term Memory; Questionnaire

Full Text:

PDF

References


Bai, X. 2018. Text classification based on LSTM and

attention. Thirteenth International Conference on Digital

Information Management (ICDIM) (pp. 29-32). Berlin:

IEEE. doi:10.1109/ICDIM.2018.8847061

Bowes, D., Hall, T., & Gray, D. 2012. Comparing the

performance of fault prediction models which report multiple

performance measures: recomputing the confusion matrix.

Proceedings of the 8th international conference on predictive

models in software engineering, (pp. 109-118).

Britz, D. 2015. Recurrent Neural Networks Tutorial, Part 1 –

Introduction to RNNs. Retrieved Maret 31, 2020, from

http://www.wildml.com/2015/09/recurrent-neural-networkstutorial-part-1-introduction-to-rnns/

Buber, E., & Diri, B. 2019. Web Page Classification Using

RNN. 8th International Congress of Information Technology

(ICITT). 154, pp. 62-72. Procedia Computer Science.

doi:https://doi.org/10.1016/j.procs.2019.06.011

Buntoro, G. A. 2017. Analisis Sentimen Calon Gubernur

DKI Jakarta 2017 di Twitter. Integer Journal, 2(1), 32-41.

Hochreiter, S., & Schmidhuber, J. 1997. Long Short Term

Memory. In Neural Computation, 9(8). Massachusetts

Institute of Technology.

Huang, W., Rao, G., Feng, Z., & Cong, Q. 2018. LSTM with

sentence representation for Document-level Sentiment

Classification. Neurocomputing, 308, 49-57.

doi:https://doi.org/10.1016/j.neucom.2018.04.045

Jozefowicz, R., Zaremba, W., & Sutskever, I. 2015. An

Empirical Exploration of Recurrent Neural Network

Architecture. International conference on machine learning,

(pp. 2342-2350). Lille.

Khurana, D., Koli, A., Khatter, K., & Singh, S. 2017. Natural

Language Processing: State of The Art, Current Threads, and

Challenges. Retrieved from ArXiv:

https://arxiv.org/abs/1708.05148

Nicholson, C. 2018. A Beginner's Guide to LSTMs and

Recurrent Neural Networks. Retrieved from Pathmind:

www.pathmind.com/wiki/lstm

Silvin. 2019. Analisis Sentimen Media Twitter Menggunakan

Long Short-Term Memory Recurrent Neural Network.

Tangerang: Universitas Media Nusantara.

Wong, T. T., & Yang, N. Y. 2017. Dependency Analysis of

Accuracy Estimates in k-fold Cross Validation. IEEE

Transactions on Knowledge and Data Engineering, 29(11),

-2427.doi:10.1109/TKDE.2017.274092


Refbacks

  • There are currently no refbacks.


Jurnal telah terindeks oleh :