“If you don’t work on important problems, it’s not likely that you'll do important work.” - Richard Hamming
In this article, we will implement and train FNet model on the Cornell Movie Dialog corpus dataset to show the applicability of FNet in text generation. read more
This article provides an introduction to cloud computing and Google cloud platform (GCP) tools and services. read more
This article provides an adaptive framework to train and serve on-device recommendation system model. This approach personalizes recommendations by leveraging on-device data, and protects user privacy without having user data leave device. read more
In this article we will examine the times series based feature extraction techniques more specifically, Fourier and Wavelet transforms. We will extract frequency and wavelet features from ECG data to train a classification model. read more
In this article, we will understand how to build NLP based text retrieval system. We will use the well-known Reuters-21578 dataset. It includes 12,902 documents for 90 classes, with a fixed splitting between test and training data. read more
In this Article we will examine a small gene expression dataset, attempting to classify leukemia patients into one of two classes. We will use some machine learning models to perform the molecular classification of cancer. read more
Many datasets in various machine learning (ML) applications have structural relationships between their entities, which can be represented as graphs. Such application includes social and communication networks analysis, traffic prediction, and fraud detection. Graph representation Learning aims to build and train models for graph datasets to be used for a variety of ML tasks. read more
EEG is a non-invasive brain imaging technique that measures difference in electrical voltage (micro-voltage), that occur as a part of neural activity in the human brain. EEG signal data acquired from subjects are unique for every individual and offers stability in signal patterns that are needed in designing biometric systems and thus have been explored as a potential biometric trait. read more
Knowledge Graphs (KGs) have emerged as a compelling abstraction for organizing the world’s structured knowledge, and as a way to integrate information extracted from multiple data sources. Knowledge graphs have started to play a central role in representing the information extracted using natural language processing and computer vision. Domain knowledge expressed in KGs is being input into machine learning models to produce better predictions. read more
Decision trees are supervised learning algorithms used for both, classification and regression tasks. Decision trees are assigned to the information based learning algorithms which use different measures of information gain for learning. We can use decision trees for issues where we have continuous but also categorical input and target features. read more
Big Data Analytics refers to the strategy of analyzing large volumes of data. This big data is gathered from a wide variety of sources, including social networks, videos, digital images, log files, sensor data, and sales transaction records, etc. The aim is to analyze all this data to discover patterns, findings, and trends that will help the concerned stakeholders to take informed decisions. read more
ANNs are nonlinear models motivated by the physiological architecture of the nervous system. They involve a cascade of simple nonlinear computations that, when aggregated, can implement robust and complex nonlinear functions. read more
The Fashion-MNIST clothing classification problem is a new standard dataset used in computer vision and deep learning. Although the dataset is relatively simple, it can be used as the basis for learning and practicing how to develop, evaluate, and use deep convolutional neural networks for image classification from scratch. read more
Principal component analysis is a widely-used dimensionality reduction technique. However, PCA has an implicit connection to the Gaussian distribution, which may be undesirable for non-Gaussian data. read more
Convolutional Neural Networks (CNNs) are the current state-of-art architecture for image classification task. Whether it is facial recognition, self driving cars or object detection, CNNs are being used everywhere. In this post, a simple 2-D Convolutional Neural Network (CNN) model is designed using keras with tensorflow backend for the well known MNIST digit recognition task. read more
Finding the efficacy and usability of a medical drug as a cure for a disease can be a problem to solve as there are chances for side effects. Generalizable solution to this problem can be achieved by exploring the post-marketing drug surveillance methods such as drug reviews analysis to monitor drug-related issues arising after taking the medication. read more
In this post we will design an algorithm to automatically identify whether a patient is suffering from breast cancer or not by analysing the histology images. read more
Data visualization is the representation of data or information in a graph, chart, or other visual format. It communicates relationships present in the data with images. read more
The objective here is to design a robust deep learning based clothing image classification tool for fashion industry. read more
In this post we will perform Market basket analysis on items purchased by customers for finding association rules to identify underlying relations between different items. read more
In this post, we’ll go through a dimensionality reduction and classification technique. LDA makes predictions by estimating the probability that a new set of inputs belongs to each class. read more