The technological know-how of this platform is inspired by the unique abilities of the human brain to analyze and solve problems. Based on the use of sensors, models, methods, algorithms and data structures, systems are developed to identify patterns, recognize objects, view countless possibilities and validate hypotheses, thus generating answers and recommendations that can help decision-making and provide adequate forecasts.
The most important know-how for this platform includes:
Technologies, methods and models to represent, store, search, evaluate, explore and retrieve knowledge, allowing inferences, predictions, abstractions and reasoning, such as ontologies, semantic Web, data mining, intelligent agents, smart scheduling, among others.
- Rule-based system
- Probabilistic reasoning
- Approximate reasoning (fuzzy logic)
- Deductive reasoning
Computer technologies, methods and tools with machine learning abilities similar to those of human beings, allowing classification and regression, prediction and clustering.
- Artificial neural networks
- Metaheuristic (e.g. particle swarm, evolutionary computation)
- Deep, statistical and never-ending learning
- Cognitive networks
Set of methods and technologies that enable computer systems to interpret images like the human eyes, allowing patterns to be recognized, information to be extracted and organized, and images to be processed, such as face biometrics, visual inspection, surveillance, process control, object recognition, medical image analysis, etc.
- Image acquisition
- Representation and description
- Mathematical morphology
- Object recognition
A set of methods and technologies by means of which computer systems are able to recognize, interpret and produce speech like human beings for device control, dialog transcription, speaker recognition and diarization, assistive speech, emotion analysis, etc.
- Speech recognition
- Voice biometrics
- Speech synthesis
- Speech Analytics
- Speech encoding
- Digital watermark
Set of methods and technologies that allow computer systems to interpret and generate dialogs like human beings, enabling the extraction and organization of information, such as emotion analysis, automatic summarization, dynamic translation, language identification, generation and understanding of natural language, etc.
Set of methods and technologies by means of which computer systems can interpret generic signs, not necessarily related to speech or vision, to recognize patterns, extract and organize information, and process signals, such as applications that involve signals from sensors, etc.