Lasmedi Afuan

Dr. Lasmedi Afuan

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 June 2021 in EDUMATIC: Jurnal Pendidikan Informatika
Verified
Journal Article
Open access
PDF file
UIUniversitas Jenderal Soedirman Program Studi Teknik Informatika
Universitas Jenderal Soedirman Program Studi Teknik Informatika
LALasmedi Afuan
Lasmedi Afuan
NNNofiyati Nofiyati
Nofiyati Nofiyati
NUNasichatul Umayah
Nasichatul Umayah
Garbage is the residual material resulting from a production process, both industry and households. Based on Brebes district's statistical data, the Paguyangan sub-district is in fourth place with an average of 249.62 m3 of garbage per day. Paguyangan is one of the villages in the Paguyangan sub-district. Paguyangan has a garbage problem where people still have the habit of throwing garbage in the river and yards around the house. Based on this, the Paguyangan Village Hall plans to build a garbage bank to overcome the problem. A garbage bank managed activities such as recording saving's transactions. This study conducted a web-based design of the Garbage Bank Information System (SIBS). SIBS is an information system used to help process garbage transactions at a garbage bank. The purpose of this SIBS is to facilitate officers and customers in processing services at the garbage bank. The methodology used in system development is the Waterfall method. SIBS development uses PHP as a programming language and MySql as the DBMS. The result of this research is an information system that can be used in the management of Waste Management transactions. The results of system testing using Black box Testing and Mean Opinion Score (MOS) obtained that all components in this system are running well according to needs.

17 January 2025 in E3S Web of Conferences
Verified
Conference Paper
Open access
PDF file
LALasmedi Afuan
Lasmedi Afuan
RIR. Rizal Isnanto
R. Rizal Isnanto
Fall detection is a major challenge in the development of technology-based healthcare systems, particularly in elderly care. This study aims to compare the performance of six classification algorithms: Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree, Random Forest, Long Short-Term Memory (LSTM), and Convolutional Neural Networks (CNN) in detecting fall incidents using wearable sensor data such as accelerometers and gyroscopes. The research utilizes a dataset consisting of 1,428 training samples and 573 testing samples, evaluated using a 10-fold cross-validation technique. The results show that CNN and LSTM deliver the best performance with accuracies of 94% and 92%, while Random Forest offers a good balance between accuracy and processing time. SVM and KNN exhibit faster processing times but slightly lower accuracies, at 87% and 84%, respectively. The superiority of CNN and LSTM in detecting more complex fall patterns aligns with previous studies emphasizing the capabilities of deep learning models in sensor data classification. The implications of these findings that the selection of fall detection algorithms must consider system priorities, whether focused on high accuracy or processing efficiency. Additionally, this research opens avenues for optimizing deep learning models and leveraging edge computing technologies to reduce response times in wearable device applications.

1 January 2024 in Journal of Applied Data Sciences
Verified
Journal Article
Open access
PDF file
LALasmedi Afuan
Lasmedi Afuan
LALasmedi Afuan
Lasmedi Afuan
Enhanced Fall Detection using Optimized Random Forest Classifier on Wearable Sensor Data

30 September 2023 in Jurnal Eksplora Informatika
Verified
Journal Article
Open access
PDF file
FDFerry Darmawan
Ferry Darmawan
MJMichael Joe
Michael Joe
YKYogiek Indra Kurniawan
Yogiek Indra Kurniawan
LALasmedi Afuan
Lasmedi Afuan
The slow handling of mental disorders and the existence of negative stigma given to individuals who have mental disorders have caused them to choose to share their problems on social media such as Twitter. This study aims to create a sentiment analysis model that is used to perform early detection of tweets from Twitter users who have signs of mental disorders in the form of depression or anxiety so that they can be handled appropriately. Sentiment analysis model created using the Support Vector Machine (SVM). The process of making a sentiment analysis model is carried out through three main stages, namely data pre-processing, SVM implementation, and result analysis. The data used is tweet data in Indonesian along with its classification label. Based on the analysis of the results that have been carried out, the best classification results by SVM are obtained using the RBF kernel at the parameter value C (complexity) = 10 and the parameter value (gamma) = 0.1, has an accuracy value of 82.5%, F1 score of 0.692, the precision value is 0.715, the recall value is 0.671, and the average F1 score is 0.637 in the 5-fold cross validation test.

1 December 2024 in Journal of Applied Data Sciences
Verified
Journal Article
Open access
PDF file
LALasmedi Afuan
Lasmedi Afuan
LALasmedi Afuan
Lasmedi Afuan
Sentiment Analysis of the Kampus Merdeka Program on Twitter Using Support Vector Machine and a Feature Extraction Comparison: TF-IDF vs. FastText

23 December 2023 in Jurnal Teknik Informatika (Jutif)
Verified
Journal Article
Open access
PDF file
AAAdrian Dwinanda Adyatma
Adrian Dwinanda Adyatma
LALasmedi Afuan
Lasmedi Afuan
EMEddy Maryanto
Eddy Maryanto
In sentiment classification systems that use Naïve Bayes Classifier, a commonly used feature extraction method is TF-IDF with unigram and bigram, where the two is used separately. In the reality, most of texts contain single or composed word,so it is needed to use the combination of unigram and bigram  to maximize the accuracy of the classification results.  In this research, the impact and performance improvement between classification systems using unigram or bigram solely and those using a combination of both are studied. Using 1000 data of reviews with ratings 1 (negative) and 5 (positive) from Gojek users on the Google Play Store, and performing performance validation with K-Fold at K=10, the system that uses the combined TF-IDF feature extraction of unigrams and bigrams achieves the best performance among the three systems with an accuracy of 0.84, however the accuracy of  the system that uses unigrams solely has accuracy of  0.83, and 0.7 for the system that uses bigram. From the results of the research, it can be concluded that the use of the combination of unigram and bigram can increase the accuracy of the classification result.