Instinctive Data Analysis in Machine Learning and Summary Exhibitor

Varshini R. S., Madhushree T., Priyadharshini R., Priya K. Y., Akshara A. S., Venkatesh J.

4th International Conference on Intelligent and Fuzzy Systems (INFUS), Bornova, Türkiye, 19 - 21 Temmuz 2022, cilt.505, ss.156-165 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 505
  • Doi Numarası: 10.1007/978-3-031-09176-6_19
  • Basıldığı Şehir: Bornova
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.156-165
  • Anahtar Kelimeler: Machine learning, Automatic data analysis, Data scientists, Classifier, Regression, 40 models, Supervised ML, Unsupervised ML
  • Yıldız Teknik Üniversitesi Adresli: Hayır

Özet

The process of Data Analysis in Machine Learning (ML) is very huge, it involves a task beginning with defining the business objective, collecting data, preprocessing the data, selecting, building and testing models, monitoring and validating against stated objectives. This requires more time for the user to get the result when each step is done manually. During analysis, not everyone checks with accuracy for all the models that exist. While dealing with ML, the data analysts usually come across lots of errors that are difficult to analyze and solve. The main objective of the paper is to perform the instinctive data analysis tool for Machine Learning in an easier way. This tool just needs the dataset, and all the data analysis required is done automatically and the result is generated within a short period. Different kinds of datasets can be provided for analysis, Eg: Numerical Dataset, Categorical Dataset, unlabelled data, etc. Around 40 regression and classifier models are available for testing here. The two main categories of Machine learning techniques have been used which are Supervised and Unsupervised. For the demo, Kaggle datasets are used, the iris dataset is used for classification, and the vegetable dataset is used for regression. This will be immensely useful for individual purposes, software companies, new budding ML engineering, and data scientists.