Skip to content

Top Headlines: Insightful Roundup of the Latest Data News

AI-Powered Predictive Analysis for Diseases and Stellar Planets: Highlights from June 22-28, 2019, featuring an innovative system that predicts the likelihood of stars hosting planets and a tool designed for beginners to swiftly produce machine learning models.

Top Trends in Data News: Its Heatering List of Updates
Top Trends in Data News: Its Heatering List of Updates

Top Headlines: Insightful Roundup of the Latest Data News

In the rapidly evolving world of technology, machine learning continues to make strides that were once thought impossible. Here's a summary of some of the most exciting recent developments in the field.

Researchers from MIT and Brown University have unveiled an innovative tool, VDS, designed to empower users with limited data science experience to generate machine learning models. This interactive system promises to democratise AI, making it more accessible to a wider audience.

Meanwhile, in the realm of astronomy, researchers from the Southwest Research Institute and several US universities have developed an algorithm that estimates the likelihood of stars hosting planets. The algorithm identifies approximately 360 stars with a greater than 90 percent probability of harbouring planets.

On the health front, a system developed by the Pentagon can analyse a person's heartbeat and identify them by their heartbeat signature from up to 200 meters away. This system, which uses lasers to detect surface movement caused by a heartbeat, can detect heartbeats through regular clothing but not thick clothing.

The VDS tool also shows promise in the medical field, as it can generate predictive models for various tasks, including predicting disease occurrence based on metabolic rate and age. Users can input datasets concerning patients' metabolic rates, ages, and disease occurrence, and the tool can predict disease occurrence based on these factors.

In the realm of neural networks, researchers have made significant strides in simulating the structural formation of the universe. A neural network, trained on 8,000 traditional universe simulations, can now create simulations with relatively few errors, a process that usually takes days but is now achievable in milliseconds.

The creators of the MLPerf benchmark suite have also made strides in the field. They have created benchmarks for image recognition, object detection, and translation. These benchmarks will test how well a machine learning system can predict a label for a given image, detect an object in an image, and translate sentences between English and German.

Lastly, the VDS tool can also generate predictive models for tasks such as predicting sales revenue. With these advancements, it's clear that machine learning is set to revolutionise various industries and aspects of our lives.

Read also:

Latest