Artificial Intelligence

In the intersection of artificial intelligence, computer science, and linguistics, this method makes machines understand human speech in text or audio format by evaluating the meaning and significance of words while completing tasks involving syntax, morphology, semantics, and discourse. By using statistical inference algorithms, it is possible for the machine to automatically learn rules through the analysis of large sets of documents. Since human speech is not always precise and often ambiguous, NLP is key in the progress of human-machine interaction applications such as virtual assistants, automatic speech recognition, machine translation, question answering, and automatic text summarization.

A method through which a computer digitizes an image, processes the data and takes some type of automated action. Machine Vision allows systems to understand and interpret the environment using live or recorded images, tag their content, and enable programs to perform automated tasks that previously required human supervision. By using one or more video cameras with analog-to-digital conversion and signal processing, a computer or robot controller receives the image data. This also allows systems to perceive their surroundings beyond regular electromagnetic wavelengths and might include infrared, ultraviolet, or X-ray frequencies to enhance image processing precision.

This machine learning method uses different algorithms to learn at multiple levels of representation and abstraction, helping to make sense of complex data such as images, sound, and text. It uses artificial neural networks to progressively extract higher-level information from the raw input. Learning can be supervised, semi-supervised or unsupervised. Deep Learning is applied in computer vision, machine vision, speech recognition, machine translation, and other areas to increase machine comprehension for inputted data.

A method of image analysis and classification based on an algorithm that classifies each pixel in a raster with the maximum likelihood of corresponding to a class. This method is widely used in remote sensing. When satellites collect information from Earth, the data is organized in a raster, or a matrix of cells, with rows and columns containing values representing information. For instance, land-use and soil data can be crossed with continuous data such as temperature, elevation, or spectral data. After performing a maximum likelihood classification on a set of raster bands, a classified raster is algorithmically created as the output according to pre-trained data. This helps to improve multi-layer remote sensing imaging and supports spatial analysis.

A data processing method that efficiently processes a large collection of satellite images to produce actionable information for decision-making procedures. By consolidating spatial, temporal, spectral, and radiometric resolution data gathered from open-access satellites, this method aims to follow land-use management in a timely and accurate manner. This is done by applying sequential image processing, extracting meaningful statistical information from agricultural fields, and storing them in a crop spectrotemporal signature library.

A 3D model-based process mounted on AI-assisted software that enables simulations from physical structures. The 3D model enables the design, simulation, and operation of what-if scenarios through virtual representations. Building Information Modelling depends on the physical and functional characteristics of real assets to provide nearly real-time simulations. The data in the model defines the design elements and establishes behaviors and relationships between model components, so every time an element is changed, visualization is updated. This allows testing different variables such as lighting, energy consumption, structural integrity, and simulating changes on the design before it is actually built. It also supports greater cost predictability, reduces errors, improves timelines, and gives a better understanding of future operations and maintenance.

This method attempts to fool machine learning models by supplying misleading inputs with the main goal of inducing a malfunction in a machine learning model. For this reason, it is deliberately used to train machine learning models in order to make them more robust and secure. Some known employed strategies are: evasion (samples are modified to evade detection), poisoning (adversarial contamination of training data), and model stealing (a black box machine learning system is used to reconstruct the model or extract the data it was trained on.)

One of the most used ways to define, quantify, and characterize the within-field variability in crop production. It is accomplished by combining geospatial data with real-time information from yield monitors mounted on combine harvesters. Variables such as mass, volume, and moisture are collected, allowing to observe both spatial and temporal yield variation within a field, helping farmers decide which parts of the area need more water or fertilizers.

A branch of Artificial Intelligence focused on developing applications or models that learn from data and enhance their performance and accuracy over time without being programmed to do so. Algorithms are trained with large amounts of data with the goal to make decisions and predictions by looking at new data. The training data is labeled to bring attention to features and classifications that a certain model will need to identify. Then, unlabeled data is presented to the model for autonomous classification, which helps to improve its accuracy. Machine Learning is widely used on digital assistants, personalized recommendations, autonomous vehicles, and robots.

A complimentary layer of machine learning that aims to achieve abstract thinking as a computational system. Machine reasoning systems are composed of a knowledge base that contains declarative and procedural knowledge and a reasoning engine that employs logical techniques such as deduction and induction to generate conclusions. The process starts with sensory and measured inputs, which gradually transform into different abstraction levels: from perceptual unstructured data, such as sensor measurements, to semi-structured and linked information, amounting to contextualized categorical descriptions of the data. This information is then transformed and fused with declarative and procedural knowledge.

Computing approach that allows machines to make sense of the real environment, translating concrete spatial data into the digital realm in real-time. By using machine learning-powered sensors, cameras, machine vision, GPS, and other elements, it is possible to digitize objects and spaces that connect via the cloud, allowing sensors and motors to react to one another, accurately representing the real world digitally. These capabilities are then combined with high-fidelity spatial mapping allowing a computer “coordinator” to control and track objects' movements and interactions as humans navigate through the digital or physical world. Spatial computing currently allows for seamless augmented, mixed and virtual experiences and is being used in medicine, mobility, logistics, mining, architecture, design, training, and so forth.

A machine translation method that applies a large artificial neural network to improve the speed and accuracy of probability predictions of a sequence of words, often in the form of sentences. Different from statistical machine translation, neural machine translation trains its parts in an end-to-end basis, performing the analysis in two stages; encoding and decoding. In the encoding stage, source language text is fed into the machine then transformed into a series of linguistic vectors. The decoding stage transfers these vectors into the target language.

The process of creating a three-dimensional representation of any surface or object (inanimate or living) via specialized software. 3D modeling is achieved manually with specialized 3D production software that allows for creating new objects by manipulating and deforming polygons, edges, and vertices or by scanning real-world subjects into a set of data points used for a digital representation. This process is widely used in various industries like film, animation, and gaming, as well as interior design, architecture, and city planning.