Conference Paper

Evaluating The Impact of Removing Less Important Terms on Sentiment Analysis

150 150 MIMOS Berhad

Authors:

Salhana Amad Darwis, Duc Nghia Pham, Ang Jia Pheng, and Ong Hong Hoe

 

Abstract:

Sentiment analysis is an important taskin Natural Language Processing (NLP) that analyses and predicts people¶s opinion from te[tualdata. It is a complex process due to the interactions with computer science, linguistics, psychology and social science disciplines. There is no straight forward rule to analyse and predict sentiment. Supervised learning methods, which adopt learning models from human, are being widely used by NLP researchers and experts to predict sentiment. However, this approach is tricky due to the challenges in ensuring the quality of the manually labelled training dataset. In this study, we investigated the use of lingXistic factors to improYe the model¶s accuracy. We gathered two datasets: (i) 125,000 annotated sentences from Amazon product reviews, and (ii) 11,250 annotated sentences from financial news articles. We then pre-processed the data, identified the less importantterms that exist in the dataset, the linguistic featuresand their effect towards the correctness of predictedsentiment. Our experimental results showed thatpunctuation separationand removal ofsupporting POS words improves precisionaccuracyin larger-generic dataset rather than in smaller-context sensitive dataset.

 

Source:

6th International Conference on Artificial Intelligence and Computer Science (AICS2019), Wuhan, Hubei, China

Benchmarking NLP Toolkits for Enterprise Application

150 150 MIMOS Berhad

Authors:

Kok Wei Ying, Duc Nghia Pham, Yasaman Eftekharypour and Ang Jia Pheng

 

Abstract:

Natural Language Processing (NLP) is an important technology that motivatesthe form of AI applicationstoday.ManyNLP librariesareavailable for researchers and developers to perform standardNLP tasks (such as segmen-tation, tokenization, lemmatization, POS tagging, and NER) without the need to develop from scratch. However, there are some challenges inselecting the most suitable librarysuch asdatatype, performance, and the compatibility.In this pa-per, we assessed five popular NLP libraries forperforming the standard pro-cessingtasksondatasetscrawled from differentonline news sourcesin Malaysia. Theobtainedresults are analysed anddifferences of those librariesarelisted. The goal of this study isto provide a clear viewfor usersto selectthe suitable NLP libraryfor their text analysistask.

 

Source:

The 16th Pacific Rim International Conference on Artificial Intelligence PRICAI 2019, Yanuca Island, Cuvu, Fiji

Study on Performance of Capacitor-less LDO with Different Types of Resistor

150 150 MIMOS Berhad

Authors:

Wan Maziyah Ab.Halim, Yuzman Yusoff, Julie Roslita Rusli, Chia Chieu Yin, and Suhaidi Shafie

 

Abstract:

This paper studies the impact of utilizing different resistor types in capacitor-less low drop-out (LDO) voltage regulator on its key performance characteristics. In order to achieve this, a 1.8 V LDO voltage regulator is designed and characterized using 180 nm CMOS technology with a supply voltage of 3.3 V. Simulations are done in schematic level using Cadence on five different types of resistor for the same LDO, and the performance in terms of output voltage accuracy, stability and power supply rejection ratio (PSRR) are compared. From the simulation results, significant differences are observedin the LDO’s performance with different types of resistor. The LDO with hpoly resistor gives the best results in terms of stability while pdiffb resistor LDO produces the highest PSRR.

 

Source:

4th IEEE International Circuit and System Symposium (ICSyS), Hotel Istana, Kuala Lumpur.

Thermal Management for High Power Devices Mount on Stack-up Printed Circuit Board (PCB)

150 150 MIMOS Berhad

Authors:

Ahmad Redzman, Khairom Nizam Mohamed, Zharfan Hamdan , Tengku Azhar, Aizuddin

 

Abstract:

To minimise foot print area of PCB in a product which all electronics components is mounted on, the PCB can be arrange in stack up condition. The electronic components comprises of high powered electronics devices such as Power Amplifier, Linear Regulator, Central Processing Unit (CPU) and Single Pole Double Throw (SPDT) switch are used in high power application such as Radio Frequency (RF) communication to boost signal power. These electronics components generate massive heat during the amplification process due to high current. This study focuses on the best ways to remove heat from these high power electronics devices in stack up PCB to the ambient in order to maintain recommended junction temperature from manufacturer. The study also discuss about ease of assembly of the stack up PCB into its casing.

 

Source:

International Journal of Electronics Communication and Computer Engineering (IJECCE), 2019

Super-Convergence of Autonomous Things

150 150 MIMOS Berhad

Authors:

Mohammad Fairus Khalid, Bukhary Ikhwan Ismail, Rajendar Kandan, ang Hong Hoe Ong

 

Abstract:

Nowadays the industry has progressed from basic mechanical assist systems to fully autonomous things such as advanced robotics, driverless vehicles and monitoring drones. The use of autonomous things is thenew revolution. In addition to IT support devicessuch as smartphones and computers we have systems that physically interact with the world and assistus with daily tasks andwork. Although there are numerous discussed benefits the realization of this autonomous things is still lacking. This paper will explore the various implementation to ease the technology adoption and proposed high level architecture for the solution.

 

Source:

The 10th International Conference on ICT Convergence; Jeju Island, Korea

Special article: Issue #3