Projects | HELA OUAILY

Shiny application for the energy sector

Sun, 25 Sep 2022 00:00:00 +0000

Creation of SQL queries to extract data from various departmental databases.
Processing and analysis of data needed to improve the planning process.
Creation, publication and documentation of an interactive dashboard using R shiny.
Classification of works and accuracy of consumption using machine learning algorithms.

The data is fictitious and the preview is fuzzy for confidential reasons.

Fall detection using Machine Learning

Fri, 22 Jul 2022 00:00:00 +0000

The aim of this project is to study the performance of several machine learning algorithms in automatically detecting falls thanks to the indicators provided. To achieve this, we will follow the following steps using Python programming: Naive Bayes- Naive Bayes Classifier – LDA : Linear Discriminant Analysis – QDA: Quadratic Discriminant Analysis – Logistic regression – KNN- K nearest neighbors – Decision Tree- Decision Tree – Random Forest– AdaBoost – Gradient Boosting.

Crypto-prediction

Wed, 22 Jun 2022 00:00:00 +0000

This is an artificial intelligence challenge I took part in on Kaggle . Using repisotory https://github.com/manthanthakker/BitcoinPrediction for inspiration, it presents implementations of machine learning algorithms (Random Forest, regression, etc.) and recurrent neural networks/long-term memory networks for BitCoin prediction. Furthermore, in our case, we have identified that BitCoin is the most important currency, as most other digital currencies will closely follow its trends. So having an accurate BitCoin prediction model should be an essential part of the project.

After importing the data obtained from Kaggle from our database, which contains historical information on several cryptocurrencies such as Bitcoin and Ethereum, I moved on to the data preparation and cleaning stage.

The next step is to deal with the missing values, rather in the difference in timestamp intervals. I started by extracting each cryptocurrency with its corresponding timestamps, visualizing them to better detect the differences and then imputing each missing value by the average of the value before and the value after.

Recognition-of-written-languages

Thu, 02 Jun 2022 00:00:00 +0000

Language modeling is a field where unsupervised learning is widely applied, with text analysis and written language detection among the best-known applications. In this project, we want to be able to identify the language of a given text (French or English). To achieve this, we’re going to use a text data table in which each text has already been labeled by its language. Initially, the aim is to build several models characterizing the different languages, based on the frequency of appearance of symbols (letters) in each language, and then to compare the different models. Each exercise in the project will involve :
¤ Choose a model
¤ Estimating its parameters
¤ Program it and comment on the results.

ShortPath

Sat, 22 Jan 2022 00:00:00 +0000

ShortPath is an R package for determining the shortest path on a graph from a source vertex using two algorithms: the Bellman-Ford algorithm and the Dijkstra algorithm.

In particular, this package can be used to determine the shortest path on a graph within the unit square, whose starting vertex has coordinates (0,0) and whose ending vertex has coordinates (1,1).

An example of the application of these algorithms is seam carving. This is an image resizing algorithm developed by Shai AVIDAN and Ariel SHAMIR, which resizes the image by removing so-called low-energy pixel paths. The package includes several functions.

See the full description by clicking the link above.

Dashboard tracking Covid-19 in Tunisia

Thu, 18 Jun 2020 00:00:00 +0000

Data

The input data for this dashboard is the dataset available from the coronavirus R package thanks to Thomas Neitmann. The data and dashboard are refreshed on a daily basis. The raw data is pulled from the Johns Hopkins University Center for Systems Science and Engineering (JHU CCSE) Coronavirus repository.

Update

The data is as of juin 18, 2020 and the dashboard has been updated on juin 19, 2020.

The dashboard was last updated in 2020, so the data is out of date.