SlideShare a Scribd company logo
1 of 3
Download to read offline
PROJECTS PORTOFOLIO
Sarcasm Detection
Clients : Dr. Afiyati, S.Si., MT.
Email : afiyati.reno@mercubuana.ac.id
Description :
In this project, Im and Dr. Afiyati, S.Si., MT. collaborated with Datains (PT Global Data Inspirasi) to
conduct research to detect sarcasm in datasets from Twitter. Create a baseline with Logistic
Regression and Naive Bayes models. Then the advanced model uses several architectural such as:
Indobertweet, Indobertweet + CNN, Indobertweet + LSTM, Indobertweet + GRU and Indobertweet +
Cosine Similarity. From the results of experiments on various architectures, the results showed that
Indobertwitter + LSTM got the highest results when compared with other architectures, including
Indobertwitter alone, with a difference of accuracy > ± 0.5%.
Improve Bag of Words with Slang Word Dictionary
Clients : Dr. Afiyati, S.Si., MT.
Email : afiyati.reno@mercubuana.ac.id
In this project, Im and Dr. Afiyati, S.Si., MT. worked together again to conduct research. Where in this
case it will try to improve the performance of a bag of words combined with an slang words dictionary
(generated from the soundex algorithm). From the experimental results it was found that a bag of
words combined with slang words and training into several algorithms such as logistic regression and
naive Bayes on average succeeded in increasing accuracy > ± 2%.
Open Source Project : POTML and POTDL Libraries
Description :
POT was the team name when Im and Dr. Afiyati, S.Si., MT. conducted joint research. This initial idea
was because we needed a flow in creating machine and deep learning models with more efficient
code and several techniques that we could develop together. The POTML library implements several
helper functions such as: statistical analysis, pipeline preprocessing, pipeline data scaling, encoding
and transformation, features selection, hyperparameter tuning, and evaluation models. POTDL
(Pytorch Base) implements several classes and helper functions such as: linear and convolutional
block layers, configuration, optimizer, training arguments for training and saving models, model
evaluation. It's as easy to use as tf.Keras but still maintains Pytorch's flexibility of architecture and
training.
Open Source Project : AutoBoosting (Automatic Machine Learning)
Description :
AutoBoosting is a complement to POTML and POTDL. Where with AutoML we can create a baseline
model quickly with just two lines of code. Which after that can automate all processes from feature
engineering, Boosting algorithm (XGBoost and LGBM) and hyperparameter tuning (random, bayes or
optuna). After the training process is complete, it will output training, validation and testing score and
provide recommendations for the best algorithm. AutoML can handle several tasks such as:
classification, regression, multioutput classification and regression, text classification, mix data types:
numeric, categorical and text for training the data.
Pretraining the BERT Model with Modification of the Attention Formula
Clients : Dr. Meilany Nonsi Tentua, S.Si, MT.
Email : meilany@ upy.ac.id
Description :
In this project, Im and Dr. Meilany Nonsi Tentua, S.Si, MT. do research together starting from
collecting data from Indonesian, English Wikipedia and Twitter for pretraining BERT model. Then the
job vacancy dataset from indeed.com (web scraping) was used for fine tuning the BERT model with a
task named entity recognition (NER). Pretraining of the BERT model was carried out on two
architectures, namely: original BERT and BERT with modified attention formulas. From the results of
fine tuning the two models, with modifying the attention formula in BERT, it succeeded in increasing
the F1 score > ± 1% compared with BERT model without modification.
Pretraining BERT Model Multilanguage (Indonesian and English)
Clients : Dr. Suprapto, M.I.Kom. (Associate Professor)
Email : sprapto@ugm.ac.id.
Description :
In this project, Im and Dr. Meilany Nonsi Tentua, S.Si, MT. do research together starting from
collecting data from Indonesian, English Wikipedia and Twitter for pretraining BERT model. Then the
review dataset of various applications from Playstore is used for fine tuning the BERT model with
sentiment analysis tasks. The review dataset from Playstore is a mixture of Indonesian and English
and this BERT model was created with the aim of being able to recognize the mixture between the two
languages . In the fine tuning evaluation results, good results were obtained in sentiment analysis with
an accuracy of 88%. In future research, we plan to create a pretrained GPT 2 model that can
recognize more than one language.
Using Narrative Context to Predict the Trajectory of Sentence Sentiment
Clients : Maya Rini Handayani, M.Kom
Description :
In this project, Im and Maya Rini Handayani, M.Kom. conducted research to predict narrative text
sentiment with a prediction range of -1 – 1. The dataset used came from the Harry Potter novel
chapters 1 – 9. Using the English BERT (Bert-base-uncased) architecture. In the experimental results,
a poor correlation value was obtained ± 0.03. In this research, there is still a lack of data, so if you
want to improve performance, the amount of data must be increased further.
GAS Detection Based On Image Using CNN
Clients : Umi Salamah, S.Si., M.Si.
Email : umi.salamah@fisika.uad.ac.id
Description :
In this project, Im and Umi Salamah, S.Si., M.Si. conducted research to detect whether an image
contains: SO2, CO2 or a mixture of both. From several experiments using pre-trained CNN models
such as: MobileNetV2 and DenseNet169. The results obtained were that the DenseNet169 model
which was trained from scratch obtained a testing accuracy of 80%.
Summarization Documents Using BERT-Kmeans
Clients : Agus Mulyanto, S.Si., M.Kom., ASEAN Eng.
Description :
In this project Im created a text summarization model using a pretrained Indonesian BERT model
combined with Kmeans clustering. BERT is used for the process of extracting paragraphs (collections
of words) into vectors. After that, a clustering process is do using Kmeans, where the paragraph
closest to the center of the cluster will be used as a representative of the document to make it more
concise.
Panorama Detection
Github : github.com/AfrizalSeptiansyah/Panorama-Detection
In this project, Im used the pretrained model “mobilenetv2” to detect objects in the panoramic image
into 5 categories: desert, mountains, sea, sunset and tree (multilabel case).
Customers Churn Prediction
Github : github.com/AfrizalSeptiansyah/Customers_Churn_Prediction
In this project, Im used the xgboost algorithm to predict whether customers will churn : yes or no
(imbalanced). Successfully found important features that managed to increase prediction accuracy.
Costumers Segmentation
Github : github.com/AfrizalSeptiansyah/Costumers-Segmentation
In this project, Im used the k-means algorithm to cluster customers into several segments, based on
similar characteristics.
Forecasting Consumer Price Index
Github : github.com/AfrizalSeptiansyah/Consumer-Price-Index
In this project, Im used fbprophet to forecast the consumer price index (cyclical problem) until the end
of 2021.

More Related Content

Similar to Portofolio Muhammad Afrizal Septiansyah 2024

Extractive Summarization with Very Deep Pretrained Language Model
Extractive Summarization with Very Deep Pretrained Language ModelExtractive Summarization with Very Deep Pretrained Language Model
Extractive Summarization with Very Deep Pretrained Language Modelgerogepatton
 
IRJET - Automatic Lip Reading: Classification of Words and Phrases using Conv...
IRJET - Automatic Lip Reading: Classification of Words and Phrases using Conv...IRJET - Automatic Lip Reading: Classification of Words and Phrases using Conv...
IRJET - Automatic Lip Reading: Classification of Words and Phrases using Conv...IRJET Journal
 
SENTIMENT ANALYSIS IN MYANMAR LANGUAGE USING CONVOLUTIONAL LSTM NEURAL NETWORK
SENTIMENT ANALYSIS IN MYANMAR LANGUAGE USING CONVOLUTIONAL LSTM NEURAL NETWORKSENTIMENT ANALYSIS IN MYANMAR LANGUAGE USING CONVOLUTIONAL LSTM NEURAL NETWORK
SENTIMENT ANALYSIS IN MYANMAR LANGUAGE USING CONVOLUTIONAL LSTM NEURAL NETWORKijnlc
 
Sentiment Analysis In Myanmar Language Using Convolutional Lstm Neural Network
Sentiment Analysis In Myanmar Language Using Convolutional Lstm Neural NetworkSentiment Analysis In Myanmar Language Using Convolutional Lstm Neural Network
Sentiment Analysis In Myanmar Language Using Convolutional Lstm Neural Networkkevig
 
Performance Comparison between Pytorch and Mindspore
Performance Comparison between Pytorch and MindsporePerformance Comparison between Pytorch and Mindspore
Performance Comparison between Pytorch and Mindsporeijdms
 
IEEE 2014 MATLAB IMAGE PROCESSING PROJECTS Scale adaptive dictionary learning
IEEE 2014 MATLAB IMAGE PROCESSING PROJECTS Scale adaptive dictionary learningIEEE 2014 MATLAB IMAGE PROCESSING PROJECTS Scale adaptive dictionary learning
IEEE 2014 MATLAB IMAGE PROCESSING PROJECTS Scale adaptive dictionary learningIEEEBEBTECHSTUDENTPROJECTS
 
Analysis of the evolution of advanced transformer-based language models: Expe...
Analysis of the evolution of advanced transformer-based language models: Expe...Analysis of the evolution of advanced transformer-based language models: Expe...
Analysis of the evolution of advanced transformer-based language models: Expe...IAESIJAI
 
IRJET- Semantic Question Matching
IRJET- Semantic Question MatchingIRJET- Semantic Question Matching
IRJET- Semantic Question MatchingIRJET Journal
 
IRJET - Mobile Chatbot for Information Search
 IRJET - Mobile Chatbot for Information Search IRJET - Mobile Chatbot for Information Search
IRJET - Mobile Chatbot for Information SearchIRJET Journal
 
IRJET - Fake News Detection using Machine Learning
IRJET -  	  Fake News Detection using Machine LearningIRJET -  	  Fake News Detection using Machine Learning
IRJET - Fake News Detection using Machine LearningIRJET Journal
 
A Research Paper on HUMAN MACHINE CONVERSATION USING CHATBOT
A Research Paper on HUMAN MACHINE CONVERSATION USING CHATBOTA Research Paper on HUMAN MACHINE CONVERSATION USING CHATBOT
A Research Paper on HUMAN MACHINE CONVERSATION USING CHATBOTIRJET Journal
 
Derogatory Comment Classification
Derogatory Comment ClassificationDerogatory Comment Classification
Derogatory Comment ClassificationIRJET Journal
 
Consumer Purchase Intention Prediction System
Consumer Purchase Intention Prediction SystemConsumer Purchase Intention Prediction System
Consumer Purchase Intention Prediction SystemIRJET Journal
 
AI-SDV 2022: Accommodating the Deep Learning Revolution by a Development Proc...
AI-SDV 2022: Accommodating the Deep Learning Revolution by a Development Proc...AI-SDV 2022: Accommodating the Deep Learning Revolution by a Development Proc...
AI-SDV 2022: Accommodating the Deep Learning Revolution by a Development Proc...Dr. Haxel Consult
 
Automatic Grading of Handwritten Answers
Automatic Grading of Handwritten AnswersAutomatic Grading of Handwritten Answers
Automatic Grading of Handwritten AnswersIRJET Journal
 
IRJET - Threat Prediction using Speech Analysis
IRJET - Threat Prediction using Speech AnalysisIRJET - Threat Prediction using Speech Analysis
IRJET - Threat Prediction using Speech AnalysisIRJET Journal
 
Graph embedding approach to analyze sentiments on cryptocurrency
Graph embedding approach to analyze sentiments on cryptocurrencyGraph embedding approach to analyze sentiments on cryptocurrency
Graph embedding approach to analyze sentiments on cryptocurrencyIJECEIAES
 
Bangla Handwritten Digit Recognition Report.pdf
Bangla Handwritten Digit Recognition  Report.pdfBangla Handwritten Digit Recognition  Report.pdf
Bangla Handwritten Digit Recognition Report.pdfKhondokerAbuNaim
 
An Intelligent Approach for Effective Retrieval of Content from Large Data Se...
An Intelligent Approach for Effective Retrieval of Content from Large Data Se...An Intelligent Approach for Effective Retrieval of Content from Large Data Se...
An Intelligent Approach for Effective Retrieval of Content from Large Data Se...IJCSIS Research Publications
 

Similar to Portofolio Muhammad Afrizal Septiansyah 2024 (20)

Extractive Summarization with Very Deep Pretrained Language Model
Extractive Summarization with Very Deep Pretrained Language ModelExtractive Summarization with Very Deep Pretrained Language Model
Extractive Summarization with Very Deep Pretrained Language Model
 
IRJET - Automatic Lip Reading: Classification of Words and Phrases using Conv...
IRJET - Automatic Lip Reading: Classification of Words and Phrases using Conv...IRJET - Automatic Lip Reading: Classification of Words and Phrases using Conv...
IRJET - Automatic Lip Reading: Classification of Words and Phrases using Conv...
 
SENTIMENT ANALYSIS IN MYANMAR LANGUAGE USING CONVOLUTIONAL LSTM NEURAL NETWORK
SENTIMENT ANALYSIS IN MYANMAR LANGUAGE USING CONVOLUTIONAL LSTM NEURAL NETWORKSENTIMENT ANALYSIS IN MYANMAR LANGUAGE USING CONVOLUTIONAL LSTM NEURAL NETWORK
SENTIMENT ANALYSIS IN MYANMAR LANGUAGE USING CONVOLUTIONAL LSTM NEURAL NETWORK
 
Sentiment Analysis In Myanmar Language Using Convolutional Lstm Neural Network
Sentiment Analysis In Myanmar Language Using Convolutional Lstm Neural NetworkSentiment Analysis In Myanmar Language Using Convolutional Lstm Neural Network
Sentiment Analysis In Myanmar Language Using Convolutional Lstm Neural Network
 
Performance Comparison between Pytorch and Mindspore
Performance Comparison between Pytorch and MindsporePerformance Comparison between Pytorch and Mindspore
Performance Comparison between Pytorch and Mindspore
 
IEEE 2014 MATLAB IMAGE PROCESSING PROJECTS Scale adaptive dictionary learning
IEEE 2014 MATLAB IMAGE PROCESSING PROJECTS Scale adaptive dictionary learningIEEE 2014 MATLAB IMAGE PROCESSING PROJECTS Scale adaptive dictionary learning
IEEE 2014 MATLAB IMAGE PROCESSING PROJECTS Scale adaptive dictionary learning
 
Analysis of the evolution of advanced transformer-based language models: Expe...
Analysis of the evolution of advanced transformer-based language models: Expe...Analysis of the evolution of advanced transformer-based language models: Expe...
Analysis of the evolution of advanced transformer-based language models: Expe...
 
IRJET- Semantic Question Matching
IRJET- Semantic Question MatchingIRJET- Semantic Question Matching
IRJET- Semantic Question Matching
 
CommitBERT.pdf
CommitBERT.pdfCommitBERT.pdf
CommitBERT.pdf
 
IRJET - Mobile Chatbot for Information Search
 IRJET - Mobile Chatbot for Information Search IRJET - Mobile Chatbot for Information Search
IRJET - Mobile Chatbot for Information Search
 
IRJET - Fake News Detection using Machine Learning
IRJET -  	  Fake News Detection using Machine LearningIRJET -  	  Fake News Detection using Machine Learning
IRJET - Fake News Detection using Machine Learning
 
A Research Paper on HUMAN MACHINE CONVERSATION USING CHATBOT
A Research Paper on HUMAN MACHINE CONVERSATION USING CHATBOTA Research Paper on HUMAN MACHINE CONVERSATION USING CHATBOT
A Research Paper on HUMAN MACHINE CONVERSATION USING CHATBOT
 
Derogatory Comment Classification
Derogatory Comment ClassificationDerogatory Comment Classification
Derogatory Comment Classification
 
Consumer Purchase Intention Prediction System
Consumer Purchase Intention Prediction SystemConsumer Purchase Intention Prediction System
Consumer Purchase Intention Prediction System
 
AI-SDV 2022: Accommodating the Deep Learning Revolution by a Development Proc...
AI-SDV 2022: Accommodating the Deep Learning Revolution by a Development Proc...AI-SDV 2022: Accommodating the Deep Learning Revolution by a Development Proc...
AI-SDV 2022: Accommodating the Deep Learning Revolution by a Development Proc...
 
Automatic Grading of Handwritten Answers
Automatic Grading of Handwritten AnswersAutomatic Grading of Handwritten Answers
Automatic Grading of Handwritten Answers
 
IRJET - Threat Prediction using Speech Analysis
IRJET - Threat Prediction using Speech AnalysisIRJET - Threat Prediction using Speech Analysis
IRJET - Threat Prediction using Speech Analysis
 
Graph embedding approach to analyze sentiments on cryptocurrency
Graph embedding approach to analyze sentiments on cryptocurrencyGraph embedding approach to analyze sentiments on cryptocurrency
Graph embedding approach to analyze sentiments on cryptocurrency
 
Bangla Handwritten Digit Recognition Report.pdf
Bangla Handwritten Digit Recognition  Report.pdfBangla Handwritten Digit Recognition  Report.pdf
Bangla Handwritten Digit Recognition Report.pdf
 
An Intelligent Approach for Effective Retrieval of Content from Large Data Se...
An Intelligent Approach for Effective Retrieval of Content from Large Data Se...An Intelligent Approach for Effective Retrieval of Content from Large Data Se...
An Intelligent Approach for Effective Retrieval of Content from Large Data Se...
 

Recently uploaded

Abortion Clinic in Kempton Park +27791653574 WhatsApp Abortion Clinic Service...
Abortion Clinic in Kempton Park +27791653574 WhatsApp Abortion Clinic Service...Abortion Clinic in Kempton Park +27791653574 WhatsApp Abortion Clinic Service...
Abortion Clinic in Kempton Park +27791653574 WhatsApp Abortion Clinic Service...mikehavy0
 
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarj
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarjSCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarj
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarjadimosmejiaslendon
 
如何办理澳洲拉筹伯大学毕业证(LaTrobe毕业证书)成绩单原件一模一样
如何办理澳洲拉筹伯大学毕业证(LaTrobe毕业证书)成绩单原件一模一样如何办理澳洲拉筹伯大学毕业证(LaTrobe毕业证书)成绩单原件一模一样
如何办理澳洲拉筹伯大学毕业证(LaTrobe毕业证书)成绩单原件一模一样wsppdmt
 
Introduction to Statistics Presentation.pptx
Introduction to Statistics Presentation.pptxIntroduction to Statistics Presentation.pptx
Introduction to Statistics Presentation.pptxAniqa Zai
 
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证pwgnohujw
 
bams-3rd-case-presentation-scabies-12-05-2020.pptx
bams-3rd-case-presentation-scabies-12-05-2020.pptxbams-3rd-case-presentation-scabies-12-05-2020.pptx
bams-3rd-case-presentation-scabies-12-05-2020.pptxJocylDuran
 
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证acoha1
 
Solution manual for managerial accounting 8th edition by john wild ken shaw b...
Solution manual for managerial accounting 8th edition by john wild ken shaw b...Solution manual for managerial accounting 8th edition by john wild ken shaw b...
Solution manual for managerial accounting 8th edition by john wild ken shaw b...rightmanforbloodline
 
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptxRESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptxronsairoathenadugay
 
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24  Building Real-Time Pipelines With FLaNKDATA SUMMIT 24  Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNKTimothy Spann
 
How to Transform Clinical Trial Management with Advanced Data Analytics
How to Transform Clinical Trial Management with Advanced Data AnalyticsHow to Transform Clinical Trial Management with Advanced Data Analytics
How to Transform Clinical Trial Management with Advanced Data AnalyticsBrainSell Technologies
 
Case Study 4 Where the cry of rebellion happen?
Case Study 4 Where the cry of rebellion happen?Case Study 4 Where the cry of rebellion happen?
Case Study 4 Where the cry of rebellion happen?RemarkSemacio
 
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi ArabiaIn Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabiaahmedjiabur940
 
Displacement, Velocity, Acceleration, and Second Derivatives
Displacement, Velocity, Acceleration, and Second DerivativesDisplacement, Velocity, Acceleration, and Second Derivatives
Displacement, Velocity, Acceleration, and Second Derivatives23050636
 
Ranking and Scoring Exercises for Research
Ranking and Scoring Exercises for ResearchRanking and Scoring Exercises for Research
Ranking and Scoring Exercises for ResearchRajesh Mondal
 
Las implicancias del memorándum de entendimiento entre Codelco y SQM según la...
Las implicancias del memorándum de entendimiento entre Codelco y SQM según la...Las implicancias del memorándum de entendimiento entre Codelco y SQM según la...
Las implicancias del memorándum de entendimiento entre Codelco y SQM según la...Voces Mineras
 
DAA Assignment Solution.pdf is the best1
DAA Assignment Solution.pdf is the best1DAA Assignment Solution.pdf is the best1
DAA Assignment Solution.pdf is the best1sinhaabhiyanshu
 
jll-asia-pacific-capital-tracker-1q24.pdf
jll-asia-pacific-capital-tracker-1q24.pdfjll-asia-pacific-capital-tracker-1q24.pdf
jll-asia-pacific-capital-tracker-1q24.pdfjaytendertech
 
Bios of leading Astrologers & Researchers
Bios of leading Astrologers & ResearchersBios of leading Astrologers & Researchers
Bios of leading Astrologers & Researchersdarmandersingh4580
 

Recently uploaded (20)

Abortion Clinic in Kempton Park +27791653574 WhatsApp Abortion Clinic Service...
Abortion Clinic in Kempton Park +27791653574 WhatsApp Abortion Clinic Service...Abortion Clinic in Kempton Park +27791653574 WhatsApp Abortion Clinic Service...
Abortion Clinic in Kempton Park +27791653574 WhatsApp Abortion Clinic Service...
 
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarj
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarjSCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarj
SCI8-Q4-MOD11.pdfwrwujrrjfaajerjrajrrarj
 
如何办理澳洲拉筹伯大学毕业证(LaTrobe毕业证书)成绩单原件一模一样
如何办理澳洲拉筹伯大学毕业证(LaTrobe毕业证书)成绩单原件一模一样如何办理澳洲拉筹伯大学毕业证(LaTrobe毕业证书)成绩单原件一模一样
如何办理澳洲拉筹伯大学毕业证(LaTrobe毕业证书)成绩单原件一模一样
 
Introduction to Statistics Presentation.pptx
Introduction to Statistics Presentation.pptxIntroduction to Statistics Presentation.pptx
Introduction to Statistics Presentation.pptx
 
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证
原件一样(UWO毕业证书)西安大略大学毕业证成绩单留信学历认证
 
bams-3rd-case-presentation-scabies-12-05-2020.pptx
bams-3rd-case-presentation-scabies-12-05-2020.pptxbams-3rd-case-presentation-scabies-12-05-2020.pptx
bams-3rd-case-presentation-scabies-12-05-2020.pptx
 
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证
如何办理(WashU毕业证书)圣路易斯华盛顿大学毕业证成绩单本科硕士学位证留信学历认证
 
Solution manual for managerial accounting 8th edition by john wild ken shaw b...
Solution manual for managerial accounting 8th edition by john wild ken shaw b...Solution manual for managerial accounting 8th edition by john wild ken shaw b...
Solution manual for managerial accounting 8th edition by john wild ken shaw b...
 
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptxRESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
 
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24  Building Real-Time Pipelines With FLaNKDATA SUMMIT 24  Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
 
How to Transform Clinical Trial Management with Advanced Data Analytics
How to Transform Clinical Trial Management with Advanced Data AnalyticsHow to Transform Clinical Trial Management with Advanced Data Analytics
How to Transform Clinical Trial Management with Advanced Data Analytics
 
Case Study 4 Where the cry of rebellion happen?
Case Study 4 Where the cry of rebellion happen?Case Study 4 Where the cry of rebellion happen?
Case Study 4 Where the cry of rebellion happen?
 
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi ArabiaIn Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
 
Displacement, Velocity, Acceleration, and Second Derivatives
Displacement, Velocity, Acceleration, and Second DerivativesDisplacement, Velocity, Acceleration, and Second Derivatives
Displacement, Velocity, Acceleration, and Second Derivatives
 
Ranking and Scoring Exercises for Research
Ranking and Scoring Exercises for ResearchRanking and Scoring Exercises for Research
Ranking and Scoring Exercises for Research
 
Las implicancias del memorándum de entendimiento entre Codelco y SQM según la...
Las implicancias del memorándum de entendimiento entre Codelco y SQM según la...Las implicancias del memorándum de entendimiento entre Codelco y SQM según la...
Las implicancias del memorándum de entendimiento entre Codelco y SQM según la...
 
DAA Assignment Solution.pdf is the best1
DAA Assignment Solution.pdf is the best1DAA Assignment Solution.pdf is the best1
DAA Assignment Solution.pdf is the best1
 
jll-asia-pacific-capital-tracker-1q24.pdf
jll-asia-pacific-capital-tracker-1q24.pdfjll-asia-pacific-capital-tracker-1q24.pdf
jll-asia-pacific-capital-tracker-1q24.pdf
 
Bios of leading Astrologers & Researchers
Bios of leading Astrologers & ResearchersBios of leading Astrologers & Researchers
Bios of leading Astrologers & Researchers
 
Abortion pills in Doha {{ QATAR }} +966572737505) Get Cytotec
Abortion pills in Doha {{ QATAR }} +966572737505) Get CytotecAbortion pills in Doha {{ QATAR }} +966572737505) Get Cytotec
Abortion pills in Doha {{ QATAR }} +966572737505) Get Cytotec
 

Portofolio Muhammad Afrizal Septiansyah 2024

  • 1. PROJECTS PORTOFOLIO Sarcasm Detection Clients : Dr. Afiyati, S.Si., MT. Email : afiyati.reno@mercubuana.ac.id Description : In this project, Im and Dr. Afiyati, S.Si., MT. collaborated with Datains (PT Global Data Inspirasi) to conduct research to detect sarcasm in datasets from Twitter. Create a baseline with Logistic Regression and Naive Bayes models. Then the advanced model uses several architectural such as: Indobertweet, Indobertweet + CNN, Indobertweet + LSTM, Indobertweet + GRU and Indobertweet + Cosine Similarity. From the results of experiments on various architectures, the results showed that Indobertwitter + LSTM got the highest results when compared with other architectures, including Indobertwitter alone, with a difference of accuracy > ± 0.5%. Improve Bag of Words with Slang Word Dictionary Clients : Dr. Afiyati, S.Si., MT. Email : afiyati.reno@mercubuana.ac.id In this project, Im and Dr. Afiyati, S.Si., MT. worked together again to conduct research. Where in this case it will try to improve the performance of a bag of words combined with an slang words dictionary (generated from the soundex algorithm). From the experimental results it was found that a bag of words combined with slang words and training into several algorithms such as logistic regression and naive Bayes on average succeeded in increasing accuracy > ± 2%. Open Source Project : POTML and POTDL Libraries Description : POT was the team name when Im and Dr. Afiyati, S.Si., MT. conducted joint research. This initial idea was because we needed a flow in creating machine and deep learning models with more efficient code and several techniques that we could develop together. The POTML library implements several helper functions such as: statistical analysis, pipeline preprocessing, pipeline data scaling, encoding and transformation, features selection, hyperparameter tuning, and evaluation models. POTDL (Pytorch Base) implements several classes and helper functions such as: linear and convolutional block layers, configuration, optimizer, training arguments for training and saving models, model evaluation. It's as easy to use as tf.Keras but still maintains Pytorch's flexibility of architecture and training. Open Source Project : AutoBoosting (Automatic Machine Learning) Description : AutoBoosting is a complement to POTML and POTDL. Where with AutoML we can create a baseline model quickly with just two lines of code. Which after that can automate all processes from feature engineering, Boosting algorithm (XGBoost and LGBM) and hyperparameter tuning (random, bayes or optuna). After the training process is complete, it will output training, validation and testing score and provide recommendations for the best algorithm. AutoML can handle several tasks such as: classification, regression, multioutput classification and regression, text classification, mix data types: numeric, categorical and text for training the data.
  • 2. Pretraining the BERT Model with Modification of the Attention Formula Clients : Dr. Meilany Nonsi Tentua, S.Si, MT. Email : meilany@ upy.ac.id Description : In this project, Im and Dr. Meilany Nonsi Tentua, S.Si, MT. do research together starting from collecting data from Indonesian, English Wikipedia and Twitter for pretraining BERT model. Then the job vacancy dataset from indeed.com (web scraping) was used for fine tuning the BERT model with a task named entity recognition (NER). Pretraining of the BERT model was carried out on two architectures, namely: original BERT and BERT with modified attention formulas. From the results of fine tuning the two models, with modifying the attention formula in BERT, it succeeded in increasing the F1 score > ± 1% compared with BERT model without modification. Pretraining BERT Model Multilanguage (Indonesian and English) Clients : Dr. Suprapto, M.I.Kom. (Associate Professor) Email : sprapto@ugm.ac.id. Description : In this project, Im and Dr. Meilany Nonsi Tentua, S.Si, MT. do research together starting from collecting data from Indonesian, English Wikipedia and Twitter for pretraining BERT model. Then the review dataset of various applications from Playstore is used for fine tuning the BERT model with sentiment analysis tasks. The review dataset from Playstore is a mixture of Indonesian and English and this BERT model was created with the aim of being able to recognize the mixture between the two languages . In the fine tuning evaluation results, good results were obtained in sentiment analysis with an accuracy of 88%. In future research, we plan to create a pretrained GPT 2 model that can recognize more than one language. Using Narrative Context to Predict the Trajectory of Sentence Sentiment Clients : Maya Rini Handayani, M.Kom Description : In this project, Im and Maya Rini Handayani, M.Kom. conducted research to predict narrative text sentiment with a prediction range of -1 – 1. The dataset used came from the Harry Potter novel chapters 1 – 9. Using the English BERT (Bert-base-uncased) architecture. In the experimental results, a poor correlation value was obtained ± 0.03. In this research, there is still a lack of data, so if you want to improve performance, the amount of data must be increased further. GAS Detection Based On Image Using CNN Clients : Umi Salamah, S.Si., M.Si. Email : umi.salamah@fisika.uad.ac.id Description : In this project, Im and Umi Salamah, S.Si., M.Si. conducted research to detect whether an image contains: SO2, CO2 or a mixture of both. From several experiments using pre-trained CNN models such as: MobileNetV2 and DenseNet169. The results obtained were that the DenseNet169 model which was trained from scratch obtained a testing accuracy of 80%.
  • 3. Summarization Documents Using BERT-Kmeans Clients : Agus Mulyanto, S.Si., M.Kom., ASEAN Eng. Description : In this project Im created a text summarization model using a pretrained Indonesian BERT model combined with Kmeans clustering. BERT is used for the process of extracting paragraphs (collections of words) into vectors. After that, a clustering process is do using Kmeans, where the paragraph closest to the center of the cluster will be used as a representative of the document to make it more concise. Panorama Detection Github : github.com/AfrizalSeptiansyah/Panorama-Detection In this project, Im used the pretrained model “mobilenetv2” to detect objects in the panoramic image into 5 categories: desert, mountains, sea, sunset and tree (multilabel case). Customers Churn Prediction Github : github.com/AfrizalSeptiansyah/Customers_Churn_Prediction In this project, Im used the xgboost algorithm to predict whether customers will churn : yes or no (imbalanced). Successfully found important features that managed to increase prediction accuracy. Costumers Segmentation Github : github.com/AfrizalSeptiansyah/Costumers-Segmentation In this project, Im used the k-means algorithm to cluster customers into several segments, based on similar characteristics. Forecasting Consumer Price Index Github : github.com/AfrizalSeptiansyah/Consumer-Price-Index In this project, Im used fbprophet to forecast the consumer price index (cyclical problem) until the end of 2021.