Three branches of artificial intelligence: cognitive, machine learning, deep learning

Artificial intelligence has entered everything – from autonomous cars to automatic emails to smart homes. You seem to get any merchandise (such as medical health, flight, travel, etc.) and make it smarter through the special application of artificial intelligence. So unless you believe that the event has a terminator-like turn, you might ask yourself if artificial intelligence can predict what benefits the workplace or the overall line of business has.

There are three main branches of artificial intelligence:

1. Cognitive AI

Cognitive computing is the most popular branch of artificial intelligence, responsible for all interactions that feel "like people." Cognitive AI must be able to handle complexity and ambiguity with ease, while continuing to learn from data mining, NLP (natural language processing) and intelligent automation.

There is a growing tendency to think that cognitive AI mixes the best decisions made by artificial intelligence with the decisions of human workers to monitor more difficult or uncertain events. This can help expand the applicability of artificial intelligence and generate faster, more reliable answers.

2. Machine Learning AI

Machine Learning (ML) AI is the kind of artificial intelligence that can automatically drive your Tesla on the highway. It is also at the forefront of computer science, but it is expected to have a significant impact on everyday workplaces. Machine learning is about looking for "patterns" in big data, and then using these patterns to predict results without too much human interpretation, and these patterns are not visible in ordinary statistical analysis.

However, machine learning requires three key factors to be effective:

a) data, large amounts of data

In order to teach new skills to artificial intelligence, a large amount of data needs to be input to the model to achieve reliable output scoring. For example, Tesla has deployed an auto-steering feature to its car, sending all the data it collects, driver interventions, successful evasion, false alarms, etc. to the headquarters, learning and correcting the senses in error. A good way to generate a lot of input is through the sensor: whether your hardware is built-in, such as radar, camera, steering wheel, etc. (if it's a car), or you prefer the Internet of Things. Bluetooth beacons, health trackers, smart home sensors, public databases, etc. are just a few of the more and more sensors connected via the Internet. These sensors can generate large amounts of data (to make it easier for any normal person to handle it). many).

b) discovery

To understand data and overcome noise, machine learning uses algorithms that can sort, slice, and transform confusing data into understandable insights. (If you want to scare off your colleagues, please listen to the different sorting algorithms that are commonly used)

The video is 5:50 in length, please watch it under WiFi conditions.

There are two algorithms for learning from data, unsupervised algorithms and supervised algorithms.

Unsupervised algorithms only process numbers and raw data, so descriptive labels and dependent variables are not established. The purpose of the algorithm is to find an inner structure that people did not expect to have. This is very useful for gaining insight into market segmentation, relevance, outliers, and more.

On the other hand, supervised algorithms use tags and variables to know the relationships between different data sets and use these relationships to predict future data. This may come in handy in climate change models, predictive analytics, content recommendations, and more.

c) deployment

Machine learning needs to go from the computer science laboratory to the software. More and more suppliers like CRM, Marketing, ERP, etc. are improving their ability to integrate embedded machine learning or to provide services that provide it.

3. Deep Learning

If machine learning is at the forefront, then deep learning is cutting-edge. This is an AI that you will send it to an intelligence question and answer. It combines the analysis of big data with unsupervised algorithms. Its applications typically revolve around large untagged data sets that need to be structured into interconnected clusters. This inspiration for deep learning comes entirely from the neural network in our brains, so it can be properly called artificial neural networks.

Deep learning is the foundation of many modern speech and image recognition methods and has higher accuracy over time than the non-learning methods previously offered.

I hope that in the future, Deep Learning AI can answer customer's inquiries and complete orders through chat or email. Or they can help market on new products and specifications based on their huge data pool. Or maybe one day they can become full-time assistants in the workplace, completely blurring the boundaries between robots and humans.

Artificial intelligence survives and improves by the scale of the data used on it, which means that not only do we see better artificial intelligence over time, but their development will revolve around organizations that can mine the largest data sets.