\u2022 Friend Suggestions:<\/a><\/li><\/ul><\/li><\/ul><\/li><\/ul><\/nav><\/div>\nThe rise of machine learning made a big splash in the year 2018. It has
<\/p>\n
Let us see the steps to master Machine Learning Technology. <\/strong><\/p>\nFor a successful Machine Learning Model, ability to train, test and validate before deploying into production is essential. Preparing data for analytics accelerates the machine learning and data science projects delivering immersive business consumer experience. This automates the data-to- insight pipeline following the below mentioned six steps.<\/p>\n\n\n
<\/span>1. Data Collection <\/span><\/h4>\n\n\n\n
The data collection step of Machine Learning is essential as the quality\nand quantity of the gathered data will determine how good the analytical model\nwill turn out to be. The multiple files need to be combined into a single input\nfile. The collected data is tabulated and named as Training Data.<\/p><\/p>\n\n\n\n
<\/span>2. Filter Data<\/span><\/h4>\n\n\n\n
The step involves loading data into a suitable place and then organise for further use. The order of the data is randomised to verify if the order does not affect the predicted results. <\/p><\/p>\n\n\n\n
<\/span>3. Analyse the data<\/span><\/h4>\n\n\n\n
The filtered data is then analysed to see if the data is suitable to be\nused for machine learning. Later, the data is divided into training and\nevaluation sets. The step deals with removing duplicates, correcting errors,\nmissing values, normalisation, data type conversions, etc.<\/p><\/p>\n\n\n\n
<\/span>4. Train the Model<\/span><\/h4>\n\n\n\n
A specific algorithm is designed to perform a particular task. The step\nis crucial as it encompasses the highly important decision to choose the\ncorrect algorithm for a model. The model is trained such that the results it\ndisplays are accurate. The aim to train the model is answering a question or\nmake predictions as often as possible. The iteration process defines every step\nof the training.<\/p><\/p>\n\n\n\n
<\/span>5. Evaluate Model<\/span><\/h4>\n\n\n\n
Performance of the model is measured using a combination of metrics. The\nmodel is tested on the previously unused data, which helps to tune the model in\na better way. <\/p><\/p>\n\n\n\n
<\/span>6. Parameter Tuning<\/span><\/h4>\n\n\n\n
Post evaluating the algorithm, improvements are made by tuning the\nparameters. It includes several training steps, learning rate,\ninitialisation values and distribution. <\/p><\/p>\n\n\n\n
<\/span>7. Make Predictions<\/span><\/h4>\n\n\n\n
Prediction is the final step where a few questions are answered. You can\nfinally detect whether the ML model can be used to predict the outcome. It\ngives the approximate result of how the model will perform in the real world. <\/p><\/p>\n\n\n\n
Commonly Used Machine Learning Algorithms<\/strong><\/p>\n\n\n\n
Moving to the digital transformation, the technology has opened the\ndoors for the tech giants to compete with each other for top data science\ntalent. The primary aim is to enable the computers to learn automatically\nwithout human intervention or assistance and adjust actions accordingly. The\ngraph of investments in technology is increasing every year. The technology has\nseveral algorithms, which can be applied to almost every data problem. <\/p><\/p>\n\n\n\n
Let us delve in deep to see the different algorithms in Machine Learning.<\/strong><\/p><\/p>\n\n\n\n<\/span>1. Linear Regression <\/span><\/h2>\n\n\n\n
Linear Regression is based on Supervised Learning algorithm of Machine Learning. The algorithm is used to estimate real values like the cost of houses, several calls, etc. based on a continuous variable(S). It is mostly used for finding the relations between independent and dependent variables by fitting the best line. This line is represented by a linear equation Y= a *X + b and is termed as a regression line. <\/p><\/p>\n\n\n\n
The model is trained using the below variables:<\/strong><\/p>\n\n\n\nX: input training data<\/p>\n\n\n\n
Y: labels to data (supervised learning) <\/p>\n\n\n\n
While training a model, to predict the value of y for a given value of\nx, it fits the best line. The best regression line is obtained by finding the\nvalues of a and b. <\/p><\/p>\n\n\n\n
a: coefficient of X<\/p>\n\n\n\n
b: intercept<\/p>\n\n\n\n
<\/span>2. Logistic Regression<\/span><\/h2>\n\n\n\n
Logistic Regression is not a regression algorithm but is a supervised classification algorithm. It helps to estimate discrete values like 0\/1, yes\/no, true\/false depending on the given set of the independent variable(s). In simple words, the output variable y can predict only discrete values for a given set of input variable x. It is also known as logit regression since it predicts the probability of occurrence of an event by fitting data to a logit function. Its output value lies between 0 and 1 as it predicts the probability. It models the data using a sigmoid function.<\/p><\/p>\n\n\n\n
Logistic Regression is classified as:<\/strong><\/p>\n\n\n\n
\u2022 Binomial: The target variable has only two possible\nvalues \u201c0\u201d and \u201c1\u201d.<\/p><\/p>\n\n\n\n
\u2022 Multinomial: Target variable has 3 or more variable.\nIt means there is no quantitative significance. <\/p><\/p>\n\n\n\n
\u2022 Ordinal: The target variables have ordered\ncategories. For example, a performance score can be categorised as: \u201cvery\npoor\u201d, \u201cpoor\u201d, \u201cgood\u201d, \u201cvery good\u201d, \u201cexcellent\u201d. <\/p><\/p>\n\n\n\n
<\/span>3. Decision Tree <\/span><\/h2>\n\n\n\n
Decision tree falls under supervised learning algorithm and is mostly used for classification problems. It is used for both categorical and continuous dependent variables. The algorithm involves tree representation where each leaf node corresponds to a class label, and the internal node of the tree represent attributes. It can be used to represent the Boolean function. <\/p><\/p>\n\n\n\n
Assumptions to be made while using a decision tree <\/strong><\/p>\n\n\n\n
\u2022 Initially, the whole training set is considered as a\nroot<\/p><\/p>\n\n\n\n
\u2022 Feature values are considered a categorical while;\nthe continuous values are discretised before building the model.<\/p><\/p>\n\n\n\n
\u2022 Records are distributed recursively based on the\nattribute values.<\/p><\/p>\n\n\n\n
\u2022 Statistical methods are used to order attributes as\nroots or an internal node. <\/p><\/p>\n\n\n\n
<\/span>4. kNN (k- Nearest Neighbors)<\/span><\/h2>\n\n\n\n
Though the algorithm is widely used to solve classification problems, it can also solve regression problems. It is a simple algorithm storing all available cases and classifying new cases by the majority votes of its k neighbours. The K nearest neighbour is measured by the distance function like Euclidean, Manhattan, Minkowski and Hamming distance. If K=1, than the case, is allocated to the class of its nearest neighbour. <\/p><\/p>\n\n\n\n
Things to consider while selecting kNN: <\/strong><\/p>\n\n\n\n\u2022 It is computationally expensive<\/p>\n\n\n\n
\u2022 Variables need to be normalised<\/p>\n\n\n\n
\u2022 Works better on pre-processing stage before moving\nto kNN outlier, noise removal<\/p><\/p>\n\n\n\n
<\/span>5. K-Means<\/span><\/h2>\n\n\n\n
K-means is a type of unsupervised algorithm used to solve the clustering problem. The procedure involves classifying the given data set through several clusters (assume k clusters). Data points within the cluster are homogenous and heterogeneous to the peer groups. Every cluster in K-means has its own centroid. Sum of the square of the difference between the centroid and the data points constitutes the sum of the square value for that cluster. Adding the sum of the square values for all clusters gives the cluster solution. <\/p><\/p>\n\n\n\n
Forming of clusters:<\/strong><\/p>\n\n\n\n
\u2022 The algorithm picks k number of points for every\ncentroid.<\/p><\/p>\n\n\n\n
\u2022 Every data point forms a cluster with the closest\ncentroid<\/p><\/p>\n\n\n\n
\u2022 Based on an existing cluster member, the centroid of\neach cluster is identified. New centroids are formed.<\/p><\/p>\n\n\n\n
\u2022 Post new centroid formation step 2 and 3 are\nrepeated. Identify the closest distance for each data point from new centroids\nand get associated with new k-clusters. Repeat the process until the centroids\ndon\u2019t change. <\/p><\/p>\n\n\n\n