Singularity Pulse  
     
  Finance

Money Management Services to Be Offered By AI

Metro Bank is planning to launch a money management service titled “Insight” for its customers using AI technologies. It makes use of predictive analytics and AI to monitor transaction data and patterns to identify trends in customers’ spending habits. It then makes personalized recommendations to them suggesting ways to reduce expenditure. It also alerts them on unusual spending, avoid additional fees, and duplicate transactions.

 
  Image  
 

OCBC Bank to Launch AI-Powered Voice Banking

OCBC Bank has launched AI powered voice banking in collaboration with Google. Customers can speak to Google Assistant on smartphone to converse about bank’s services. Services include planning for retirement, home loans and education loans. Customers can also ask for mortgage loan amount they can avail. Foreign exchange rates and financial market updates are provided to them upon requests. OCBC is first bank in Singapore to collaborate with Google on such kind of service platform.

Banking  
  Image  
  Information Technology

An Algorithm to Locate Fake Users on Social Networks

Researchers at University of Washington & University of the Negev have come up with new method to detect fake accounts on social networking sites like Facebook and Twitter. The algorithm works on two iterations. First, it constructs a link prediction classifier that can estimate probability of a link between two users. Second iteration generates a new set of meta-features based on the features created by the link prediction classifier. Researcher uses meta-features and constructed a generic classifier that can detect fake profiles on platforms.

 
  Image  
 

AI to Guide on Earthquake Response

One Concern, startup in California, is using ML and AI to advise fire departments on how to plan for earthquakes and respond to them. They feed their computers three categories of data. First is the kind of material used in building homes, second data point is on natural environment and third is live instant data, like magnitude of earthquake occurred. Algorithm then makes predictions. It also uses past data to measure how good its predictions were and improve on them simultaneously.

Information Technology  
  Image  
  Defense

US Army Uses AI for Recognizing Faces Even in Dark

U.S. Army researchers have developed a new facial recognition technology which uses A.I. and a person’s heat map to recognize them. It uses FLIR Boston 320 thermal camera as it takes person’s heat map and compares it to a biometric database to find a match. Biometric database includes data from criminal records and watch lists. Researchers claim that visible technology is not as effective as thermal technology. The algorithm focuses on people’s features like eyes, nose, and mouth to make recognitions. The technology is called “convolutional neural network.”

 
  Image  
 

ML to Reveal Cause-Effect Relationship in Protein Dynamics Data

Researchers at University of South Florida have integrating machine learning techniques to develop an algorithm which can identify cause-effect relationships in proteins. Proteins carry out specific tasks through various stimuli, such as small molecules. Where proteins carry out their specific tasks is controlled by cells through various stimuli, such as small molecules. Stimuli interact with proteins and modify their speeds and strengths. Machine learning algorithm will enable them to find cause-effect relationships in protein dynamics data.

Research  
  Image  
  Public Policy

AI to Scour Code to Find Accidentally Public Passwords

Researchers at Pivotal, a software infrastructure firm, have trained their AI powered algorithm to locate sensitive information which is accidentally kept public. Information like passwords etcetera at public hotspots qualify for such incidences. The algorithm does it by looking at the code as if it were a picture. It identifies difference between a password and normal code. To speed up the entire process, text of password and code is converted into lists of numbers describing each string of characters. The software is trained of deep learning algorithm.

 
  Image  
 

Skin Care Goes into The Hands Of AI

Olay, the global skincare brand, has developed an AI powered Skin analysis application called, “Olay's Skin Advisor”. It can provide with skin assessment by analyzing the selfie taken by the person. The algorithm also asks short questionnaire about one’s skin concerns to tailor made the recommendations. From its launch date, around 3.8 million women have already used the platform for getting personalized recommendations and solutions for their skin worries.

Fashion  
  Image  
  Entertainment

AI Algorithm to Tell What Music Is Playing in Your Mind

According to a research carried out at the University of California, Berkeley, an app can now identify the music going in your brain by simply analyzing the thought patterns. Brian Pasley used deep-learning algorithm on brain activity to turn thoughts into digitally synthesized speech. Person’s brain waves were analyzed while they were speaking. Team recorded a pianist’s brain activity when he played music, then performed the experiment again asking the pianist to imagine the notes instead. This training was used to create their music-predicting algorithm.

 
  Image  
 

Data Science and ML to Brew Beer

Tyler Hutcherson, data science engineer and Metis Machine, teamed up with Champion Brewing Company, a local brewery to develop first-ever beer brewed using data science and ML. They are calling it ML IPA. The brewing company provided them with parameters upon which the IPAs is judged in Beer Festival. They matched the range with 10-best-selling and 10 worst selling IPAs at local stores. The resultant IPA has 6% alcohol by volume (ABV), 60 IBU (bitterness scale) and 6 SRM (color system).

Miscellaneous  
  Image  
 
 
 
     
  Banner  
 
Contact us:

5A, B-Wing, Trade Star Building, J. B. Nagar, Andheri-Kurla Road, Andheri (East), Mumbai - 400 059, Maharashtra, India
W: www.decimalpointanalytics.com  
E: info@decimalpointanalytics.com   T: +91 22300 15200