AI's Case Study|NTT DATA Philippines

Case Study

Achievement of AI utilization widely spread from knowledge discovery to prediction, automation / autonomization

08/10/2018

NTT Data aims to provide “AI that responds to business issues”.

We are advancing the use of a series of AI technology such as rule base, machine learning, deep learning for all utilization scenes such as knowledge discovery, prediction, and automation.

Digital_AI12

Case study 1 [Automation]: Automatic generation of news manuscript
Considering the possibility of “AI Reporter” realization

Challenge

In the media industry, the move toward automatic generation of news articles is accelerating. However, many existing technologies generate articles by embedding words and numerical values in prepared template statements, and we rely on human for design. For this reason it was a time-consuming process to apply to multiple fields.

Solution

In collaboration with business persons in Japanese media industry, NTT DATA has initiated a demonstration experiment on technology to automatically generate weather news documents with relatively simple patterns. We built a mechanism that weather cloud data for the past 4 years released by the Meteorological Agency and weather news script actually read out by the announcer are set and learning is done by deep learning. As a result of evaluating weather news manuscripts generated from this method, Japanese grammar has reached a level that does not cause discomfort even when people read it, and some correction is necessary in the correctness of meaning, but we have confirmed that we can create documents with the same contents as weather telegrams in general.

Furthermore, we are planning to proceed with new demonstration experiments in fields that involve massive amounts of data, such as corporate financial results announcements and sports articles.

Case study 2 【Discovery】: Smart ICU
Predictive detection of deterioration of seriously ill patients

Challenge

Medical practices in the Intensive Care Unit (ICU) are specifically targeted at severe patients, so they are performed while monitoring patient conditions at all times using sophisticated medical equipment. However, integration with IT systems has not progressed so much, such as enormous data management being performed manually. There were cases that took a lot of time for physicians and nurses to interpret data along with various test results and medication information.

Solution

NTT DATA and the EVERIS group of the Spanish subsidiary in collaboration with Virgin Del Rossio University Hospital have centralized management of all data related to patients by consolidating information obtained from various medical devices in the ICU into one platform. Based on various data including vital accumulated and accumulated by this, we developed a model that predicts the risk of complications to occur 2 hours before onset by AI technology.

Utilizing such a model for predicting the symptomatic transition by AI, when the system predicts the occurrence of complications of a patient, it can immediately notify the bedside terminal and the mobile terminal of the risk, and the doctor and the nurse can confirm it "smart alert Solution "developed. At the same time, doctors can make quick diagnosis on the spot by simultaneously providing vital data necessary for case diagnosis at the time of notification.

Case study 3 [Automation]: Congestion alleviation
Congestion prediction / signal control simulation

Challenge

In Guiyang City, the provincial capital of Guizhou Province, China, due to economic growth and urbanization progress, traffic congestion in the city center is a major social problem.

Solution

NTT Data collaborated with the Chinese Academy of Sciences Software Research Laboratory to analyze large-scale traffic data collected through cameras for traffic management. We conducted a demonstration experiment to control about 220 traffic lights at the intersection of 19 Lake Yama Lake District in Guiyang City, China, by AI to optimize signal parameters by conducting congestion prediction / signal control simulation. As a result of considering the effect of alleviating congestion and improving the traffic throughput at the intersection, we confirmed that traffic congestion in the target area improved by 7% on average, 26% in maximum, and the traffic processing volume improved by an average of 6.7%.

Case study 4 [Automation]: Watch over the elderly by robots and sensors
Consider the possibility of improvement of care efficiency in aging society

Challenge

The population of Japan is aging rapidly, and in 2060 it is predicted that 26.9% of the total population will be over 75 years old. While the number of people who need nursing care increases as the population aging progresses, the number of nursing staff is on the shortage trend. Measures are needed to reduce the workload of nursing care staff while providing adequate care to care recipients.

Solution

Digital_AI12

Transition of AI technology

In collaboration with a social welfare corporation, NTT DATA conducted a demonstration experiment to monitor the elderly and record their daily life in nursing care facilities using communication robots and sensors. We have considered the ability to alleviate heavy care at the nursing care workplace and reduce burden on nursing care work by detecting wake-up at night with a risk of falling by warning of the robot by voice or checking health at the time of getting up in the morning.

Furthermore, we are planning to expand the scope of application to home care as well as proceed with demonstration experiments to provide more accurate watching functions and evaluate the reduction effect of nursing care work burden.

We are advancing the use of a series of AI technology such as rule base, machine learning, deep learning for all utilization scenes such as knowledge discovery, prediction, and automation.

You may also like