scholarly journals Unsupervised machine learning for exploratory data analysis in imaging mass spectrometry

2020 ◽  
Vol 39 (3) ◽  
pp. 245-291 ◽  
Author(s):  
Nico Verbeeck ◽  
Richard M. Caprioli ◽  
Raf Van de Plas
Keyword(s):  
Machine Learning ◽  
Mass Spectrometry ◽  
Data Analysis ◽  
Exploratory Data Analysis ◽  
Imaging Mass Spectrometry ◽  
Unsupervised Machine Learning ◽  
Exploratory Data

scholarly journals Understanding Question Pair Similarity

2021 ◽  
Vol 9 (VI) ◽  
pp. 2632-2638
Author(s):  
Shekkari Akhil
Keyword(s):  
Machine Learning ◽  
Data Analysis ◽  
Natural Language ◽  
Exploratory Data Analysis ◽  
Feature Engineering ◽  
Modelling Strategies ◽  
Exploratory Data ◽  
Natural Language Process

One of the most important areas where the Natural Language Process of Machine Learning may help is determining if two questions are similar. The model we create can instantly detect if a question is similar to one that has already been posed. To find the underlying patterns in our data, we'll do a complete Exploratory Data Analysis. Based on our observations, we will do feature engineering. We'll try out a few different modelling strategies to determine which one works the best and keeps the greatest outcomes.

scholarly journals Rekomendasi Fitur yang Mempengaruhi Harga Sewa Menggunakan Pendekatan Machine Learning

2021 ◽  
Vol 8 (4) ◽  
pp. 673
Author(s):  
Bambang Wisnuadhi ◽  
Irwan Setiawan
Keyword(s):  
Machine Learning ◽  
Data Analysis ◽  
Exploratory Data Analysis ◽  
Machine Learning Techniques ◽  
Web Based ◽  
Tourist Attractions ◽  
User Data ◽  
Exploratory Data ◽  
Tourism And Hospitality Industry ◽  
Tourism And Hospitality

<p class="abstrak">Perkembangan Teknologi Infromasi, internet, dan perangkat bergerak telah mengubah perilaku konsumen dalam menjalankan aktivitasnya. Hal ini direspon oleh industri dengan menyediakan berbagai aplikasi berbasis web dan perangkat bergerak dalam interaksinya dengan pelanggan. Salah satu industri yang beradaptasi dengan perubahan teknologi dan perilaku konsumen ini adalah industri pariwisata dan perhotelan. Kebutuhan konsumen yang sebelumnya menggunakan akomodasi wisata tradisional seperti hotel, berubah menjadi lebih memilih rumah-rumah penduduk disekitar tempat wisata sebagai tempat penginapan sementara wisatawan. Perubahan ini berdampak kepada semakin banyaknya properti pribadi yang disewakan sehingga menyebabkan persaingan harga sewa. Harga sewa merupakan salah satu faktor penting yang dipertimbangkan calon penyewa dalam menentukan properti yang akan disewanya. Hal ini tentunya membuat para pemiliki properti harus memikirkan strategi penentuan harga sewa agar propertinya laku dipasaran. Penelitian ini bertujuan untuk mendapatkan fitur apa saja yang dapat mempengaruhi penentuan harga sewa properti berdasarkan data pengguna Airbnb di Berlin. Data penelitian diambil dari dataset yang disediakan oleh InsideAirbnb berupa file dengan format CSV. Penelitian dilakukan menggunakan teknik <em>machine learning</em> dengan pendekatan algoritma XGBoost. Terdapat lima tahapan pengerjaan dalam penelitian ini, yaitu <em>data understanding, data pre-processing, exploratory data analysis</em>, pemodelan, dan <em>insights</em>. Hasil yang didapatkan dari penelitian ini adalah <em>room type private room, room type entire home/apt</em>, dan <em>cancellation policy super strict 60 days</em> merupakan tiga fitur tertinggi yang mempengaruhi penentuan harga sewa. Luas properti menempati urutan keempat berdasarkan rekomendasi algoritma yang diterapkan.</p><p class="abstrak"> </p><p class="abstrak"><em><strong>Abstract</strong></em></p><p class="abstrak"><em>The development of information technology, the internet, and mobile devices has changed the behavior of consumers in carrying out their activities. The industry responded by providing various web-based and mobile applications in their interactions with customers. The tourism and hospitality industry is adapting to changes in technology and consumer behavior. The needs of consumers who previously used traditional tourist accommodations such as hotels have changed to prefer residents' houses around tourist attractions as their temporary lodging. This change has an impact on the increasing number of private properties being leased, causing competition in rental prices. It is undeniable that the rental price is one of the essential factors that prospective tenants consider in making choices. This certainly makes property owners, who will rent out their properties, have to think about rental pricing strategies. This study aims to obtain any features that affect pricing based on Airbnb user data in Berlin. The study was conducted using machine learning techniques with the XGBoost algorithm approach. There are five stages of work in this study, namely understanding data, pre-processing data, exploratory data analysis, modeling, and insights. The results obtained from this study are room type private room, room type entire home / apt, and cancellation policy type super strict 60 are the three highest features that affect price determination. Property size ranks fourth based on algorithmic recommendations.</em></p><p class="abstrak"><em><strong><br /></strong></em></p>

scholarly journals Exploratory Data Analysis and Machine Learning Algorithms to Classifying Stroke Disease

2021 ◽  
Vol 2 (02) ◽  
pp. 77-82
Author(s):  
Prismahardi Aji Riyantoko ◽  
Tresna Maulana Fahrudin ◽  
Kartika Maulida Hindrayani ◽  
Mohammad Idhom
Keyword(s):  
Machine Learning ◽  
Data Analysis ◽  
Gradient Descent ◽  
Exploratory Data Analysis ◽  
Learning Algorithm ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Stochastic Gradient Descent ◽  
Machine Learning Algorithm ◽  
Exploratory Data

This paper presents data stroke disease that combine exploratory data analysis and machine learning algorithms. Using exploratory data analysis we can found the patterns, anomaly, give assumptions using statistical and graphical method. Otherwise, machine learning algorithm can classify the dataset using model, and we can compare many model. EDA have showed the result if the age of patient was attacked stroke disease between 25 into 62 years old. Machine learning algorithm have showed the highest are Logistic Regression and Stochastic Gradient Descent around 94,61%. Overall, the model of machine learning can provide the best performed and accuracy.        

Sign in / Sign up

Export Citation Format

Share Document