Gartner has a magic quadrant to show you who is the best company in a certain environment. As you can see here, Gartner has this chart here called a Hype Cycle. They have a hype cycle for various environments like data management and so on but today we’re going to look at the hype cycle for emerging technologies. So what it shows here is within this hype cycle they’ll show what are the up-and-coming technologies in the next 2 – 5 years times and even up to 10 years.

In 2017, let’s look at the bottom side we start, there are a lot of technologies and there are up & coming. With all of these, we’re going to focus on one environment which is being repetitive within the next few years. The topic we talking about here is Artificial Intelligence (AI). Even back in 2017, AI has already been in the hype cycle. We start with Artificial General Intelligence. It might take more than 10 years for them to trigger. More than 10 years just to trigger the start of AI but if you look further up, we have more than that, we have an IoT platform, an Internet of thing platform as well, deep learning, and machine learning. These are up and coming within 2 to five years’ time. We’ll learn this later on, but just to show you what we have in the hype cycle. In 2017, we even have blockchain here. Blockchain is the most popular cryptocurrency. Cryptocurrency has exploded in terms of market size. Bitcoin has gone up to 50 000 US dollars per coin. 

We move forward to the following year 2018, you’ll see the repetitive item come back up with Artificial General intelligence. Now, we’re moving deeper portion, we have HAI, AI platform, and IoT platform. The trend is moving toward this field similar item AI and IoT.

In 2019, we’re moving nearer and this is 2 years past already, here very simple, the white dot basically is more than 10 years and the yellow dot is just 5 – 10 years. We can see, now we’re going into a deep dive portion of AI and now we have adaptive machine learning. For slightly higher, we have even emotional AI, explainable AI, HAI, and so on. There are more and more AI technologies than one general AI. Now we are branching out to multiple parts of AI.

In 2020, you can see within the next 2 – 5 years time what we have. We do have AI-Augumented design, we go slightly higher same things which are adaptive machine learning, generative AI, composable AI, responsible AI, and explainable AI. We’re no longer just looking at one path from one AI path we can move toward multiple paths itself.

Lastly, in 2021, everything is repeated here. From 2017, we have 10 years of incoming technology, and now 5 years have reached where are we right now within AI. There’re more technologies are coming up. We’re actually in the digital era where a lot of our environment has either integrated a small portion of AI or they are using it as a whole as well to run a fully automated system. From the above chart, we have AI-Augumented design, physics Informed AI, and slightly higher we have Generative AI. Those are the same things. Besides that, last year and this year, we have Nonfungible Tokens (NFT). 

What is NFT? NFT people can just sell a piece of art for millions of dollars just from Nonfungible Tokens. These are up-and-coming technologies. As a person who is new to IT, when do we want to venture? When you want to venture into IT, where are you supposed to start? What type of skill set do you want to equip yourself with? Do you want to include yourself in programming languages? Do you want to go with technology and learn more about blockchain, AI, and NFT? As you can see, AI is definitely one of the trends that will continue for the next few years to come within 5 – 10 years. 

What is AI? What is Artificial Intelligence?

AI is one thing that we want to use to simulate human intelligence into machines. 

According to one of the very famous American psychologists, the name is Howards Gartner. He said that human Intelligence can be classified into 7 categories such as visual intelligence, Linguistic Intelligence, and verbal intelligence so there are 7 categories to which one of the examples here of human intelligence is as I mentioned Linguistic Intelligence. Linguistic Intelligence is the intelligence that can allow us as a human to be able to understand speech and ability to express ourselves, to express our own ideas by using the right words. As mentioned earlier, AI is one thing that we want to use to simulate human intelligence into machines. One of the examples is we can simulate Linguistic Intelligence into the machine so the machine can also identify speech and respond with the right word. For example, the effort series, Microsoft Cortana and etc. 

Visual Intelligence is one of the Intelligence of humans that can allow humans like us and we are able to perceive the environment. From what we perceive, we’ll be able to identify objects nearby us. For example, laptop in front of me and etc. When this visual intelligence of humans can be stimulated into the machine, we can actually build a system to which you can refer the Autonomous vehicle which has the capability to identify the word condition by seeing through the smart camera and drive safely on the road. So, basically, it is a very brief introduction to what is AI. Ai is to simulate human intelligence into machines. 

When we talk about AI, one of the very popular terms is machine learning. What is the relationship between machine learning and AI? Machine learning is not equal to AI but is one of the tools that we can use to build the AI system. It is not at the same level as AI but it is a subset of AI.

Important concept of machine learning: Decision-making System

What machine learning will do, we’ll look into one of the very important concepts here which is the decision-making system. What is a decision-making system? So, we as a human, are using our Intelligence to make tons of decisions every single day. Every decision our make, we make is based on our own decision-making system. For example, if we want to decide where is this animal or not, first, we will try to ask our decision-making system, we will try to examine whether this animal has an ear, have 4 legs, eyes, or a tail, then we only will make a conclusion that this is a cat. So, as a human, we will always have our own decision-making system which is actually a set of rules to be referred to before we can make any decision.

Important concept of machine learning: Decision-making System 

Knowledge to identify something but as a human, if we want to take action, for example, if we want to win a basketball game, we also need to refer to our decision-making system to decide which is the best action to be made in every different scenario in the game in order to win the game. For example, when the scenario is if somebody is blocking me from going near the basket, the decision to make is whether I need to turn my body a little bit or go the right by 2 steps or etc in order to escape from that person who is blocking me. Besides that, another scenario will be I need to make a shoot at the position maybe 3 feet away from the basket. Then the decision that I have to make in order to make a good shoot is that maybe I need to bend my knee a little bit, raise the ball with the use of a suitable force, a right angle of 40 degrees to throw the ball only I can make the successful shoot. We are actually doing a lot of decision-making every day, and it will all be based on the decision-making system that we learned here.

Machine Learning is a study of model/algorithms that formulates rules for machines.

The reason why I need to tell you about decision-making systems is that machine learning is the study of models or algorithms that formulates the rules for machines. In machine learning, we want this machine learning model to learn decision-making systems. As a human, how can e derive a decision-making system? We learn from our experiences because every day we are going through different things and different events. So from day to day, from time to time, we are gaining experience. The more experience we can gain, the better will be the decision-making rules that we can derive. When we have better decision rules, the better decision we can make, so that will be how we as a human can derive our own rules. In the machine learning process, we want to learn the machine learning model with the data so the data is like the experience to humans. We’ll need to fit good data into the model so the model will come up with its own rules. Machine learning also tries to derive good rules from the data so when they need to make some decisions or make some predictions will be referring back to their decision-making system to give us a good prediction. So this is basically a machine learning process, we are going to train the model to derive a good decision-making system from the data.

How to train a machine to cheer up a girl?​

Last example here, the goal of the machine is to cheer up the girl. How can this machine know what are the better steps to be taken in order to achieve that goal? We’ll need to train the machine so that it can derive its own rules-making system. During the machine learning process or the training process. Different kinds of information can be perceived by the machine. For example, if the robot tries to give the snake to the girl or the girl will be very unhappy. However, if the robot gives flowers to the girl and the girl will be very happy. If the robot gives the girl some money so the girl will be even happier. So based on every data information here, the machine will update its own rule-making system so it will remember what should be done and what shouldn’t be done. After proper training here, the machine will get the final rule-making system which it will refer to if they go right and refer back to this decision-making system that it learned during the machine learning training.

Machine learning as mentioned earlier would be just one of the subsets of Artificial Intelligence. It is just one part of the AI. The machine learning model is just like the brand of our AI system well to obtain a completely functional AI system. We still need to get other components sensors so we’re going to get some smart cameras or the internet of things to access the hand or other part of the air model that can collect data and maybe help us display the result from the model processing. As we said AI systems often need to process a huge volume of data especially in a single short time so it is also very important that we can learn about big data because big data will ensure that all these processing the machine a massive amount of data can be completed in the required time. 

AI is just a small portion of your to-do processing of data. We’re setting a preset rule so that our machine will be able to process the data and know what to do next. As a human, we have a brain to do that, whereas a machine doesn’t have a brain to do that. What does the machine need to do to be able to learn we set a preset rule such as 1+1=2 and 2+2=4.

Now, where do we get the rules set, and where do we collect all this data, that’s where IoT comes in. IoT will act as your hand, eyes, nose, ears, and mouth. So you’ll be able to hear, taste, smell, and everything. As mentioned earlier, AI is just one small subset, you can build up an entire full ecosystem to run which we called a full automate’s item as well. That’s where you will get IoT in AI, and a big data to run. 

Why are these 3 items work so well together? IoT is responsible to collect data, from your hand, eyes, nose, and ears (collecting data, feeling, seeing, and smelling). However, AI will be processing data such as holding a cup of hot water, showering in cold water, seeing my favorite food, and smelling it. AI is helping you to process, it is like your brain is processing, and understanding what the object that has here that are being collected from IoT devices. Lastly, big data will help you to analyze all of this. For example, you might see your favorite(you like it). AI helps you to process that’s a bowl of noodles or a plate of chicken rice maybe that’s some of your favorite food. AI will only tell you this is a plate of chicken rice, this is a vegetable or this is a steak. Your brain will analyze based on the past experience that you have experienced. When we have a lot of data being gathered, our human brain won’t be able to process all of it at once as well. 

How do we study? For example, we try to memorize by repetition when we have so much data that we try to memorize and try to repeat. The way to go through with this is big data to be able to analyze to help you to make an informed decision. For example, e-commerce in Malaysia, such as Shopee and Lazada. How do you think Shopee and Lazada do their advertising? Whenever new users sign up for Shopee and Lazada, what will you see on the first page whenever you log in? They usually show you what the up-and-coming products are or what products are popular. Where do there get all of this information? These are based on users’ clicks. 

Shopee and Lazada have millions of users every month especially since they have monthly promotions that might be even higher so whenever users click on certain items, these are where they generate data, one click equals one data. Once you click on ID products, ID products get one point. You click into kitchenware or household items or you click into any item every individual system will get one point, that’s where IoT will be collecting, AI will be processing one point higher than another, and the second point will be higher than the third one. There will be able to analyze which product has a higher click, higher view, or even higher product being sold. 

Once you process, big data can come in to help you to do precision marketing. There can do more precise advertising toward targeted users. For example, 10 products here, are top-selling products but they are from different backgrounds, some are technology, sports equipment, kitchen item, clothing brand, and so on. So, how do we know what the items look like? That’s where analyzed, what the user look past or item purchase or view. Mainly based on IT technology, when promoting the 10 items, only promote 10 IT products to the user.

When you have all of these items, IoT will help you to collect your data, like your arm, eye and etc to help you to collect. AI is like your brand, it is to help you to process. They will help you to process what are the items. Once you analyze the user’s previous behavior when you can have a fully automatic system to help you to gather your data to process it and analyze. Just a quick understanding of what AI is and where can you apply it. There is much technology we can go through as well, especially what we saw through the Ghana Cycle, you can apply some of those skill sets into your environment and reuse those skill sets to venture into a field of IT