Fariz Ihsan Yazid Blog

started on week 10 our team tries to code machine learning with the help of XGBoost algorithms, somehow it’s difficult and according to the documentation it is very unusual for us, in the end, we decided just tried to code.

moving on to week 11 which our team almost gave up with the XGBoost and the day of presentation is near which makes us need to find a simpler algorithm, then we found out that Linear Regression and Support Vector Machine is easier than XGBoost. even though it was easier, we still got trouble finishing our project since both algorithms show a similar result.

next, in week 12 which is the finalization of our project we tried to ask the lecturer, Ms. Nurul, to help our time supervise the project. after the project reviewed by the lecturer, we conclude that the problem was on the datasets and it wasn’t the code.

last is week 13 which the day of presentation, the day was after the new years break then we present our project in front of our lecturer and friends and while we demonstrate our projects it’s work finely and then we finish the presentation without any problem occurs.

On week five our team still doing some research on algorithms that are going to use, then our team starting to confused because each of us did not familiar with the meaning of stock detail such as (price, open, close, high, low, and volume) data.

next, on week six we already knew what stocks detail, then the next problem came into us which were the other terms that very strange to us the adj close and open price, etc. This problem needs to be solved then we research more on how to predict stock data with simple logic.

In week seven our team felts that we already knew how to predict the stock price by saw the trend, also we decide to not make it into machine learning since it requires more algorithms on the machine learning, and also we used the library rather than created from scratch since its simple.

The on week 8, we continue our research by reading lots from medium.com, towarddatascience.com,and many article about predictions of stock. After spend lots of time to research and reading, we decide to use Random Forest Algorithms and at the end our team still unable to code since lack of understanding andd choosing algorithm randomly.

On week nine we finally found an article from https://towardsdatascience.com/https-medium-com-vishalmorde-xgboost-algorithm-long-she-may-rein-edd9f99be63d and it show an image that explainning lots of algorithms here is the picture:

Then we decided to use xgboost since it has lot references.

For this semester I learn about Intelligent System which this course is associated with AI and Machine Learning. this course was lectured by sir Ford Lumban Gaol and ms N. Nurul Qomariah with the help of lab assistants Mr. Andreas Kurniawan then I will explain a little bit about our progress in this page.

Week 1 – 4

First, on week one I need to form a team in order to do the final project then I formed up with David and Frendy, also our lab assistants asked us to decide what are we going to create however until week two we didn’t have a clue what we are going to create. In week two we decided to create a stock prediction and also we consults this idea with sir Andreas and he gave us the feedback which later on we stick to this project.

Second, from week three to four we didn’t do anything yet for the final project, we only do research to find any algorithms that suit our project.