17 Unsupervised Learning Interview Questions (SOLVED) To Brush Before Data Science Interview

Some common challenges that unsupervised learning can help with are:

• Insufficient labeled data: For supervised learning, there is a requirement for a lot of labeled data for the model to perform well. Unsupervised learning can automatically label unlabeled examples. This would work by clustering all the data points and then applying the labels from the labeled ones to the unlabeled ones.
• Overfitting: Machine learning algorithms can sometimes overfit the training data by extracting too much from the noise in the data. When this happens, the algorithm is memorizing the training data rather than learning how to generalize the knowledge of the training data. Unsupervised learning can be introduced as a regularizer. Regularization is a process that helps to reduce the complexity of a machine learning algorithm, helping it capture the signal in the data without adjusting too much to the noise.
• Outliers: The quality of data is very important. If machine learning algorithms train on outliers (rare cases) then their generalization error will be lower than if they are ignored. Unsupervised learning can perform outlier detection using dimensionality reduction and create solutions specifically for the outliers, and separately, a solution for the normal data.
• Feature engineering: Feature engineering is a vital task for data scientists to perform, but feature engineering is very labor-intensive, and it requires a human to creatively engineer the features. Representation learning from unsupervised learning can be used to automatically learn the right type of features to help the task at hand.

• Supervised learning examples:

  • You get a bunch of photos with information about what is on them and then you train a model to recognize new photos.
  • You have a bunch of molecules and information about which are drugs and you train a model to answer whether a new molecule is also a drug.
  • Based on past information about spams, filtering out a new incoming email into Inbox (normal) or Junk folder (Spam)
  • Cortana or any speech automated system in your mobile phone trains your voice and then starts working based on this training.
  • Train your handwriting to OCR system and once trained, it will be able to convert your hand-writing images into text (till some accuracy obviously)

• Unsupervised learning examples:

  • You have a bunch of photos of 6 people but without information about who is on which one and you want to divide this dataset into 6 piles, each with the photos of one individual.
  • You have molecules, part of them are drugs and part are not but you do not know which are which and you want the algorithm to discover the drugs.
  • A friend invites you to his party where you meet totally strangers. Now you will classify them using unsupervised learning (no prior knowledge) and this classification can be on the basis of gender, age group, dressing, educational qualification or whatever way you would like. Why this learning is different from Supervised Learning? Since you didn't use any past/prior knowledge about people and classified them "on-the-go".
  • NASA discovers new heavenly bodies and finds them different from previously known astronomical objects - stars, planets, asteroids, blackholes etc. (i.e. it has no knowledge about these new bodies) and classifies them the way it would like to (distance from Milky way, intensity, gravitational force, red/blue shift or whatever)
  • Let's suppose you have never seen a Cricket match before and by chance watch a video on internet, now you can classify players on the basis of different criterion: Players wearing same sort of kits are in one class, Players of one style are in one class (batsmen, bowler, fielders), or on the basis of playing hand (RH vs LH) or whatever way you would observe and classify it.

• Neural Networks are used in unsupervised learning to learn better representations of the input data.
• Neural networks can learn a mapping from document to real-valued vector in such a way that resulting vectors are similar for documents with similar content. This can be achieved using autoencoders which is a model that is trained to reconstruct the original vector from a smaller representation with reconstruction error as a cost function.
• There are neural networks that are specifically designed for clustering as well. The most widely known is the self-organizing maps (SOM).

• As the amount of data required to train a model increases, it becomes harder and harder for machine learning algorithms to handle. As more features are added to the machine learning process, the more difficult the training becomes.
• In very high-dimensional space, supervised algorithms learn to separate points and build function approximations to make good predictions.

When the number of features increases, this search becomes expensive, both from a time and compute perspective. It might become impossible to find a good solution fast enough. This is the curse of dimensionality.

• Using dimensionality reduction of unsupervised learning, the most salient features can be discovered in the original feature set. Then the dimension of this feature set can be reduced to a more manageable number while losing very little information in the process. This will help supervised learning find the optimum function to approximate the dataset.

• K-nearest neighbors or KNN is a supervised classification algorithm. This means that we need labeled data to classify an unlabeled data point. It attempts to classify a data point based on its proximity to other K-data points in the feature space.

• K-means Clustering is an unsupervised classification algorithm. It requires only a set of unlabeled points and a threshold K, so it gathers and groups data into K number of clusters.

• Supervised learning is when the data you feed your algorithm with is tagged or labelled, to help your logic make decisions.

  Example: a hypothetical non-machine learning algorithm for face detection in images would try to define what a face is (round skin-like-colored disk, with dark area where you expect the eyes etc). A machine learning algorithm would not have such coded definition, but would "learn-by-examples": you'll show several images of faces and not-faces and a good algorithm will eventually learn and be able to predict whether or not an unseen image is a face.

• Unsupervised learning are types of algorithms that try to find correlations without any external inputs other than the raw data (your examples are not labeled, i.e. you don't say anything). In such a case the algorithm itself cannot "invent" what a face is, but it can try to cluster the data into different groups, e.g. it can distinguish that faces are very different from landscapes, which are very different from horses.

• k-Means Clustering is a method of vector quantization, that aims to partition n observations into k clusters in which each observation belonging to the cluster with the nearest mean serving as the prototype of the cluster.
• Different runs of k-means clustering result in different cluster assignments because k-means randomly assigns each observation to one of the k clusters to kick off the clustering process. Typically, the k-means algorithm does several runs and chooses the run with the best separation, defined as the lowest total sum of within-cluster variations across all k clusters.
• Data points with which k-means clustering has been done is shown in the figure below:

• As part of their construction, random forest predictors naturally lead to a dissimilarity measure among the observations. One can also define a random forest dissimilarity measure between unlabeled data:

  • the idea is to construct a random forest predictor that distinguishes the observed data from suitably generated synthetic data.

• Many unsupervised learning methods require the inclusion of an input dissimilarity measure among the observations. Hence, if a dissimilarity matrix can be produced using Random Forest, unsupervised learning can be successfully implemented. The patterns found in the process will be used to make clusters.

• LLE goes beyond the density modeling of the PCA. Density models do not provide a consistent set of global coordinates which embed the observations across the entire manifold. These can only detect linear patterns in the data and does a bad job detecting the curved pattern. LLE is able to detect the curved path of the dataset as shown below:

• LLE is more efficient compared to other algorithms in terms of its computation space and time because it tends to accumulate sparse matrices.

• Unsupervised learning deals with clustering and associative rule mining problems.
• Reinforcement Learning deals with exploitation or exploration, Markov’s decision processes, Policy Learning, Deep Learning and value learning.

• Unsupervised learning deals with unlabeled data where the output is based on the underlying patterns of the data.
• Reinforcement learning deals with the Markov's Decision process where the agent interacts with the environment in discrete steps, and the data is not predefined.

• Unsupervised learning discovers underlying patterns and predicts the output.
• In Reinforcement Learning, the learning agent works as a reward and action system by following a trial and error method.

Hierarchical Clustering is a method of cluster analysis which seeks to build a hierarchy of clusters. The hierarchical clustering algorithm is run without a preconfigured number of clusters, then after the clusters are formed, the designer can pick and choose what clusters are required. The two types of hierarchical clustering are:

• In the Agglomerative clustering, each observation starts in its own cluster and pairs of clusters are merged as one moves up the hierarchy.
• In the Divisive clustering, all the observations start in one cluster, and splits are performed recursively as one moves down the hierarchy.

• A hierarchical clustering algorithm builds a dendrogram which makes it easier to understand the algorithm. Dendrograms of the two types of hierarchical clustering are shown in the figure below: