Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal IJIES (International Journal of Innovation in Enterprise System) Indonesian Journal of Electrical Engineering and Computer Science
Naomie Salim
Faculty of Computer Science and Information System Universiti Teknologi Malaysia
Computer Science & IT Control & Systems Engineering

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
A joint learning classification for intent detection and slot filling with domain-adapted embeddings Muhammad, Yusuf Idris; Salim, Naomie; Zainal, Anazida
Indonesian Journal of Electrical Engineering and Computer Science Vol 37, No 2: February 2025
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v37.i2.pp1306-1316

Abstract

For dialogue systems to function effectively, accurate natural language understanding is vital, relying on precise intent recognition and slot filling to ensure smooth and meaningful interactions. Previous studies have primarily focused on addressing each subtask individually. However, it has been discovered that these subtasks are interconnected and achieving better results requires solving them together. One drawback of the joint learning model is its inability to apply learned patterns to unseen data, which stems from a lack of large, annotated data. Recent approaches have shown that using pretrained embeddings for effective text representation can help address the issue of generalization. However, pretrained embeddings are merely trained on corpus that typically consist of commonly discussed matters, which might not necessarily contain domain specific vocabularies for the task at hand. To address this issue, the paper presents a joint model for intent detection and slot filling, harnessing pretrained embeddings and domain specific embeddings using canonical correlation analysis to enhance the model performance. The proposed model consists of convolutional neural network along with bidirectional long short-term memory (BiLSTM) for efficient joint learning classification. The results of the experiment show that the proposed model performs better than the baseline models.
Combined Approach for Teachers’ Evaluation Aspects Identification Using Dictionary and Patterns Based Rengiah, Parimala; Kaewyong, Phuripoj; Salim, Naomie; Phang, Fatin Aliah
International Journal of Innovation in Enterprise System Vol. 3 No. 2 (2019): International Journal of Innovation in Enterprise System
Publisher : School of Industrial and System Engineering, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Teacher performance evaluation is a common method and often used for evaluates teaching quality in higher education. With the rapid growth of opinion mining technique. Aspect-based opinion mining application has been possibly employed to extraction and summarization of students' comments for teacher evaluation. However, to automated teacher evaluation features identification from a large number of students' comments collection is very hard work. This study has the goal to address this problem. The main objectives of the proposed method are: (1) to identify teacher evaluation aspects, (2) to compare the efficiency of dictionary based, patterns based and the combination of them, and (3) to enhance the accuracy result in the teachers’ evaluation aspects identification from the unstructured text of students' feedbacks. The students' feedbacks were collected by questionnaires and the dataset was constructed manually with a total of 4,496 sentences from 300 undergraduate student responses in 10 subjects by purposive sampling and the collection of positive and negative sentences from 30 participants group interviewed in the workshop. Both approaches were applied to identify the frequency teachers' evaluation aspects. The experimental results found that our proposed approach provided reasonably more accurate results, the combination approach enhanced a significantly average of precision and recall. For future work, we focus on the application of new linguistic patterns and non-frequency aspects in order to increase the accuracy result. Keywords—aspects identification, lexicon relation, linguistic pattern, opinion mining, teacher evaluation.
Co-Authors Fatin Aliah Phang Kaewyong, Phuripoj Muhammad, Yusuf Idris Rengiah, Parimala Zainal, Anazida