Market Prediction with Artificial Intelligence. Demonstration.

        We finally managed to actualise what We wanted to make. 

        We developed A model that can predict the stock market and automatically notify the result to people through Twitter. 

        Once the backtesting script is ready, We can check whether or not the model is trustworthy. Thus, the backtesting programme is under development.

        At the moment, I do not think this model can precisely predict the market. We badly need to test this model.

        We described the method used to create this model in the former articles in this blog. So if you are interested, check them out.

        After ensuring the information in the source code we reveal, we will upload the source code and the weights we use in the demonstration on Github :  https://github.com/LulutasoAI. 

Failure No.331 A Stock price Chart-Pattern detection with Yolo. Part1.

        Many people utilise chart patterns for their decision-making process in their investments. The question here is can we automate a chart-pattern recognition process?

        If possible, we can examine, for example, the correlation between chart patterns and future prices.

        We used YOLOv5 for that purpose. 

        We attempted twice, and both failed. 

        First, we tried to detect chart patterns from pictures of candlestick charts.

        We made 69 labelled training data and 5 test data. It did not detect anything at all after training for 200 epochs.

        The first picture below is a labelled picture. Note that the first picture is a picture from the validation data I labelled, not the prediction that AI made.

        The second is the prediction result. No pattern is detected.

       presumably, it was because a candlestick chart with Moving Average indicators on it might be too complex, and the information it contains was redundant for the purpose. 

     Second, we changed the training pictures into 2D line charts and labelled 44 pictures. It does not, however, detect any patterns again after training. Well, usually, in this case, we might need more data to make the model learn correctly. 

        We will prepare more data for the YOLO to learn. 

        Also, we have been developing stock/index prices analytic scripts in python here https://github.com/LulutasoAI/FInancial_Analysis. we will explain the script and its usage in the following article. Stay tuned.

Hi everyone. It is advancing paw by paw!

I plan to develop an algorithm to forecast the stock market from scratch. To achieve that goal, I need to acquire more Mathematical and Statistical knowledge. It might take time. 

Although this field is drastically changing and growing daily, I think it is the right approach to be patient for now. 

 The strategy is below.

I will read articles and papers, on Machine Learning or Financial prediction with ML, on Arxiv and the Internet. I will gain insights from them and implement them continuously in my algorithm while studying Maths and Statistics. 

I am going to sophisticate the candlestick chart analysis using CNN. I also intend to actualise a Neural Network model to estimate the future prices of financial assets that utilises variables made of sentimental information extracted from Twitter or other social media. 

      One more thing, I will also do something with my English writing skills since the more I write in English, the more I realise how unsophisticated I am at it. I appreciate your patience. Thank you. 

New project: an AI democratic congress

       We have built a market decision-making algorithm in which three AIs are trying to guess whether or not the stock market will go up. 

What if we pack various, but with the same purpose, AIs into a virtual congress and let them determine the best trading decisions? We are working on it. 

Although generating AIs takes a lot of computational resources and time, an estimation process does not deplete those as much as in the process of learning. Therefore, in theory, we concluded that we could assign quite a fair number of AIs to the assembly to have virtual debates. 

Things I have learnt. I am thinking about turning the lessons into a contribution to society.

  I have been working for a company where we offer solutions using AI technologies. I keep thinking that I am not contributing enough for what I earn. However, what I have done might be more than okay for a first-year AI developer. 

I have no intention of quitting the job, but I am perhaps ready to think about creating my own business. Luckily or not, our Japanese government is willing to lend capital with relatively generous interest rates to business owners. It honestly reminds me of the financial crisis in 2008 with many subprime loans. It might, however, be a chance for me to take since macro risks like that will be taken care of by the government and not individuals like me for now. What is more, I will do the right thing with the capital. 

So I will make plans as CEO of ANEKOSYSI going to prevail soon. We will change the world into a better and more sustainable place.

Come and Join AI field!

Humanity has invented almost everything in the physical world. This sentence sticks inside of my brain every single time I come up with a seemingly new idea which turned out to have already been found out by a person.

  In the AI field, however, there is still space to explore. After working in the AI sector for four months, I realized that there are many things to automate with this brilliant technology. What is more, this field is mainly idea-based.

 Although nobody usually trusts your AI model without mathematical and statistical facts supporting it, you can create one for personal use, thanks to Keras's python library. Besides, mathematics does not matter anymore when your model does a great job. Results are pretty much everything.

 I will restart developing the market prediction AI machine with you. Although you will get a better answer at another place online, any questions are welcome.

Hello All!

I am pleased to announce that I am going to make my own business called "ANEKOSYSI". Even in Japanese, my native tongue, the name is weird enough. We needed, however, a certain amount of uniqueness in our name first. Then the philosophy is hoped to live in our unique and innovative behaviour and products.

I wanted a business named "Neko System". The name turned out to be too long and not rhythmical enough. What is more, a former insightful entrepreneur used the name already. It turned out to be Okay to me since it sounds like "Analysis", which we might do in our business. Let's make it NekoSys and sandwich it between "AI", which is our primary strength to contribute to society.

It is ANEKOSYSI's first step with its paw. Let's go