MACHINE Anomaly detection

Unsupervised learning based data sampling learning modeling and key feature extraction

Development of an optimal decision boundary search algorithm for classifying and determining normal or abnormal features

Improvement of transfer learning model

Sound recognition & generation

(CNN +)deep learning modeling to build superb audio classifier/generator to recognize every sound

Deep Learning-based generative & discriminative modeling

Object detection & recognition

Reliable object detection and recognition help robots to react adequately for quickly adapting to changes

General object(e.g. car) or robotics (e.g. skeleton) real-time detection & tracking

Multi-modal Data analysis

Multi-modal data analysis including emotion detection, heterogeneous data convergence make robot to diagnose current status with prediction

Raw source data (IoT, text, image, video, etc.) identifier management and status monitoring with ensemble approach

Cooperation Process to build A.I solution for You

1. Needs summary and pre-analysis experiment from R-SPIRIT based on research

2. Proposal to go to next step for realization

3. Co-working with data collecting-preprocessing-analysis-testing

4. Cyclic improvement by continuous market monitoring for launch