AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |
Back to Blog
Basic data of science12/29/2023 ![]() It is fairly common to use advanced techniques in competitions. Use these techniques when the requirement is to get the best accuracy. Deep learning is used for speech recognition, board games AI, image recognition, and manipulation.Īdvanced machine learning techniques give the highest accuracy. Use deep learning when the number of variables (columns) is high. Deep Learningĭeep learning is a subset of machine learning that works similar to the biological brain. Similarly, when a user watches a movie from a series, the video hosting application recommends other movies from the series. So when a user adds a pizza base to the cart, the mozzarella cheese is recommended to the user. Mozzarella cheese and pizza bases are frequently brought together. When one of the frequently bought, items is added to the cart, the e-commerce application would recommend other products in the group. The algorithm takes previous orders as input and outputs the association rules. For example, e-commerce applications use association rule mapping for finding the groups of products that are frequently bought together. In data science terminology the co-occurring patterns are known as association rules. Association Rule Mapping (Semi-Supervised)Īssociation rule mapping is the Identification of co-occurring patterns. In case of a valid reason like a tourism trip the transaction is accepted or else it is declined. The customer care of the bank may call or send a message to the user to verify the transaction. This is an unexpected event and a red flag is raised. Suppose a credit card transaction of an American citizen happens in India. One of the common use cases of anomaly detection is identifying fraudulent card transactions. Similarly, we can predict whether a movie that would be released would be a blockbuster based on its genre, movie length, profits of previous movies of the directors and actors.Īnomaly detection is the Identification of unexpected events. For example, predict whether a student would pass an English exam by checking the grammar of assignments written by him. In classification, the category of observation is predicted. In linear regression, the label variable is expressed as a linear equation of other variables. Another example of regression is to predict the volume of trees with tree height and tree girth.Īmong regression techniques, linear regression is the most popular technique. For example, predict the stock price using the previous day’s price, market movement, and news. Use regression for understanding the relationship between data and for predictions. In regression, a numerical variable is predicted using other variables. Supervised, Unsupervised, and semi-supervised learning are further classified into five types based on the work done by the algorithms. Machine learning is broadly classified into three types – Learning the various categories of machine learning, associated algorithms, and their performance parameters is the first step of machine learning. Whereas in machine learning, the algorithm understands the data and creates the logic. In conventional programming, the programmer understands the business needs, data, and writes the logic. It is a branch of artificial intelligence. Machine learning is the science of building models automatically. Both R (ggplot2) and python (Matplotlib) have excellent graphing capabilities. For exploratory data analysis use graphs and statistical parameters mean, medium, variance. Exploratory Data AnalysisĮxploratory data analysis is analyzing and understanding data. Python can be added to the skill set later. R being a statistical language is an easier option. it is overwhelming to learn data science concepts and a general-purpose language like python at the same time. For academics and domain experts, R is the preferred language. Programmers can start with either R or Python. These data models predict outcomes of new data. Since the field covers such a vast array of services, data scientists can find a ton of great opportunities in their field.ĭata scientists use algorithms for creating data models. ![]() Some of the applications of data science are driverless cars, gaming AI, movie recommendations, and shopping recommendations. Building Profile Earn Kaggle medals Publish Articles Build Resumeĭata science is analyzing and predicting data, It is an emerging field. Time series Natural Language Processing (NLP) Sentiment analysis Image recognition 8. Advanced Machine Learning Boosting Ensembling 6. Basics of Machine Learning Supervised Unsupervised Reinforcement Semi-Supervised Learning Regression (Supervised) Classification (Supervised) Clustering (Unsupervised) Anomaly Detection (Semi-Supervised) Association Rule Mapping (Semi-Supervised) 4. The Fundamentals Mathematics Linear Algebra Probability Probability distributions Statistics Programming Language (R or Python) 2. ![]()
0 Comments
Read More
Leave a Reply. |