Overfitting occurs when our machine learning model tries to cover all data points more than the required data points in the given dataset.It is because of this the model starts catching noise and inaccurate values present in the dataset and all these factors reduce the accuracy and efficiency of the model.The overfitted model has low bias and high variance.The chances of occurence of overfitting increase as much as we provide training to our model. Overfitting is the main problem that occurs in supervised learning.

Logistic regression is unsupervised machine learning that is used to solve classification problems.The classification problem has a result of 0 or 1, yes or no , true or false ,spam or not spam.

It is a predictive analysis algorithm which works on the concept of probability.

Logistic regression is a type of regression but it is different from linear regression algorithm

It uses sigmoid function or logistic function which is a complex cost function. The sigmoid is used to model in logistic regression.

The function can be represented as

Data scientists involve reinforcement learning as a multistep process which have clearly defined rules.Data scientists program an algorithm to complete a task and give it positive or negative cues as it works out to complete a task. But for most times algorithms define a rule to what steps take along the way.

How does reinforcement learning work?

Reinforcement learning works by programming an algorithm with a distinct role that is beneficial to the ultimate goal and avoid punishment.

Reinforcement learning is often used in areas such as

Robotics: Robotics can learn how to perform tasks.

Videogame Play : Teach the bots how to play different games

Resource Management : Given finite resources and a defined goal , reinforcement learning can help to plan out how to allocate resources.

This approach involves a mix of 2 preceding types.Data Scientist may feed an algorithm mostly labelled training data, but the model is free to explore data on its own and develop its own understanding of the dataset.

How does semi-supervised machine learning work?

Semi- supervised machine learning works on feeding a small amount of labelled data to an algorithm. From this the algorithm learns the dimensions of the dataset.But labelling data can be time-consuming and expensive.Some areas where semi-supervised machine learning is included.

Machine Translation:Teaching algorithms how to translate language based on less than a full dictionary of words

Fraud Detection: In case of identifying fraud , you can have only a few positive examples.

Labelling Data:Algorithms of small datasets can learn to apply datasets to a larger algorithm.

Unsupervised machine learning do not require data to be labelled .They sniff through unlabelled data to look for patterns that can be used to group data points . Unsupervised Machine Learning are good for the following points.

- Clustering : Splitting the dataset into groups based on similarity.

2. Anomaly Detection: Detection of invalid or wrong absurd datapoints in a dataset.