Saya ucapkan terimakasih kepada sobat-sobat blogger yang telah menyempatkan diri untuk membuka laman ini, blog ini saya buat sebagai media sharing kepada semua anak bangsa yang ingin belajar-belajar dan belajar khususnya mengenai ilmu statistika, saya pribadi mengucapkan mohon maaf jika terdapat kesalahan maupun kekeliruan dalam hal penyajian informasi kepada para pengunjung.
Seluruh Isi dari Blog ini ditujukan untuk PENGETAHUAN SEMATA... Tutorial yang dimuat ini untuk level NEWBIE (alias PEMULA), bukan untuk level MAS[TER]. Penyalahgunaan tutorial ataupun hal yang lainnya BUKAN Merupakan Tanggung Jawab Admin.
Ini adalah sedikit perkenalan dari saya, semoga jalinan silaturrahmi antar kita semua dapat terjalin dengan baik. Terimakasih.
- Nama Lengkap.
M. Nashihun Ulwan
Kelayu, 6 Sya'ban 1414 H
Selong - Lombok Timur - NTB
M Nashihun Ulwan
MI NW Aik Ampat
MTs. N. Model Selong
MAN Selong (IPA 1)
Universitas Islam Indonesia (Statistika)
Playing Computer [Browsing, Troubleshooting PC, Design Grafis, Programing]
Fishing and adventure.
Never Put Off Until Tomorrow What Can Do Today
You Can If You Think You Can
Utamakan Logika daripada Emosi
I am Statistician who has interested in data mining, machine learning and efficient algorithms. Working as a team of data scientist is very interesting for me. The nature of my degree has prepared me for this position. It Involved many activities to prepare myself to being a good Data Scientist. I also want to learn how to manage and the process of “big” research especially about big data. I’m very passionate and have the enthusiasm and determination to ensure that i make a success of it.
- Lab Instructor for Programming Algorithms, Database, Management Information System, Time Series Analysis, Applied Multivariate Statistics at Department of Statistics Islamic University of Indonesia.
- Internship at PT SAS Institute Jakarta
- Surveyor at KOMINFO
- Surveyor at Danamon Peduli
- Surveyor at Findyr Application
- Surveyor at Committee of the National Economy and Industry
- Publisher at Google Adsense
- Chief of Islamic Science and Research Club, FMIPA UII
- Head of Research and Development Department, Ikatan Keluarga Statistika FMIPA UII
- Staff of Research and Development Department, Lembaga Eksekutif Mahasiswa FMIPA UII
- Committee for International Conference on Statistical Methods in Engineering, Science, Economy, and Education
STATISTICS AND DATA MINING PROJECTS
- An Artificial Neural Network for Earthquake Magnitude Prediction and Spatial Clustering with Equal Size
Modeling movement points of Indonesian earthquake and predict occurrence using backpropagation neural networks which have 3 input layer, 2 hidden layers, and 1 output layer. Cluster analysis that is spatial clustering with equal size is also used to classify points of earthquake based on location, so the results of analysis can be used to determine policies and prepare the maximum mitigation efforts to reduce the risk of earthquakes.
- Market Basket Analysis using Apriori Algorithm of Association Rules
This research is tried to find patterns for items that are frequently purchased together from database of purchase transactions. The results from this research can be used to arrange the items products are often purchased together, strategize sales and marketing (promotion), the layout of items, and design discounts for purchases the product package.
- Segmentation Level of Poverty and Influencing Factors Using Self-Organizing Maps (SOMS) Kohonen and Ordinal Logistics Regression
The purpose from this paper was to determine the characteristics and factors affecting poverty in East Java. Segmentation poverty using Self-Organizing Maps (SOMS) Kohonen algorithms, The result is four groups of the poverty level are high, medium, low, and rich, then from the group conducted ordinal logistic regression analysis to determine which factors significantly affect the level poverty in each group.
- Social Media Analysis
Collecting data about GOJEK tweet from Twitter, text mining techniques are used to find words that are important and makes a wordcloud from it, find the text associations to get information and perform clustering text and sentiment analysis using machine learning method that it Support Vector Machine and Naive Bayes Classifier to determine the public perception of GOJEK.
- Killing Case Analysis From News Sites Based On Rough Set TheoryThe purpose from this paper was to show the rules incidence of killing cases that occurred, the method used is the rough set theory, an uncertain event will be reduced to obtain simpler the rules to describe a pattern of rules occurred. The attibutes that used in this research are victim, killer, cause for condition and category cilling for consequences. The results obtained rules that can be used to predict the incidence of killing cases.
- Spatial Error Model (SEM) Method For Determine the Factors of Affecting the Poverty in East Java
Factors that affect the problem of poverty in East Java be known using Spatial Error Model, by geting the model of SEM can be used to predict poverty in East Java, the results obtained that the factors affects poverty in East Java is IPM, TPAK, and Population Density
- Vehicle Insurance Claims Analysis Using Binary Logistic Regression Method using SAS Enterprise MinerBinary logistic regression analysis was used to determine the factors that cause a potential customer to make a claim against its vehicle, so the company can determine the chances of a prospective customer to make claims in the future based on factors generated. From 22 variable, there are 15 variable that significantly affect the claim are bluebook, car type, car use, education, home value, income, occupation, kids drive, marital status, driving history, single parent, revoked, TIF, distance workplaces dan urbanicity
- Pattern Recognization on Unstructured Text Data Using Support Vector Machine and Association
Text data from public reports on LAPOR! site are classified into 3 category (complaints, aspirations, questions). The classification process using Support Vector Machine method, after the classification process finished, text association generated to extract the information on each category to obtain the hidden of information from it. The results in each category can represent a topic that is often voiced by society.
M Nashihun Ulwan [OneszAccess Corporation Inc.]