Housing Index in Tunisia increased to 147.60 points in the fourth quarter of 2021 from 146.70 points in the third quarter of 2021. This dataset provides - Tunisia Housing Index- actual values, historical data, forecast, chart, statistics, economic calendar and news.

Real estate market trends, property prices, rental yields, and market forecasts

Motivation

Due to the lack of datasets that contain information about Tunisia, I have decided to start a series of datasets covering various constructive Tunisian data; starting with this house pricing data set. The creation of this dataset is an attempt to provide helpful information on house pricing in Tunisia, covering thousands of online listings provided with a considerably fair amount of information on each listing. A rather interesting aspect of the data is its geographic accuracy, this enables a lot of creative analysis opportunities.

Context

This dataset is the result of scraping "various" house sales websites. It contains more than 8000 observations with 25 attributes (excluding the id attribute). All records correspond to a house, an apartment, or a villa within the country of Tunisia.

All data was scraped in October 2022 and was processed to provide accurate data. It serves as an excellent introduction to implementing machine learning since it requires little to no data cleaning, a considerable size of data, and an understandable list of attributes.

You will find bellow two versions of this dataset one containing rows missing the prices and one with only the rows where the price is available.

Post-scraping Data modification

This dataset includes one feature that was defined by me with the attribute name of distance_to_capital. This extra feature aims to calculate the distance of a house listing to the capital Tunis given that the Latitude and longitude coordinates are, respecitvely 36.806389, 10.181667.

The distance was calculated using the Haversine forumale:

\[ d=2 r \arcsin (\sqrt{\sin ^2(\frac{ \varphi 2- \varphi 1} {2}) + \cos ( \varphi 1 ) \cos (\varphi 2) \sin ^2 ( \frac { \lambda 2 - \lambda 1 } {2} )})) \]

\(\small \rm with:\)\(\small \textrm{d: distance to be calculated.}\) \(\small \textrm{r: the radius of the Earth (6371km).}\) \(\small \varphi 1 \varphi 2: \textrm{the Latitudes of point 1 and point 2, respectively}\)\[\small \lambda 1 \lambda 2: \textrm{the Longitudes of point 1 and point 2, respectively}\]

Content

The columns are as follows, their names are pretty much self-explanatory:

• id: 128-bit universally unique identifier.
• price_tnd: the price of a given listing in Tunisian Dinars.
• price_eur: the price of a given listing in Tunisian Euros. This was converted on 29 October 2022 with a conversion rate of 1 TND equals 0.31 euros.
• location: the neighborhood of a given listing.
• city: the city of a given listing.
• governorate: the governorate of a given listing out of 24 Tunisian governorates.
• area: the area of a giving listing in square meters.
• room: the number of rooms.
• bathroom: the number of bathrooms.
• age: the age of a listing provided in intervals with the value 0 indicating it's new.
• state: the state of a listing provided in 3 different values; 1 indicating it's in a normal state, 2 indicating it requires renovation and 0 indicating it's brand new.
• latt: latitude.
• long: longitude.
• distance_to_capital: the distance between a listing and the capital Tunis.
• garage: boolean, indicating the existence of a garage.
• garden: boolean, indicating the existence of a garden.
• concierge: boolean, indicating the existence of a concierge.
• beach_view: boolean, indicating whether the listing has a beach view or not.
• mountain_view: boolean, indicating whether the listing has a mountain view or not.
• pool: boolean, indicating the existence of a pool.
• concierge: boolean, indicating the existence of a concierge.
• elevator: boolean, indicating the existence of an elevator.
• furnished: boolean, indicating whether the listing is furnished or not.
• equipped_kitech: boolean, indicating whether the listing kitchen is equipped or not.
• central_heating: boolean, indicating whether the listing has central heating or not.
• air_conditionting: boolean, indicating whether the listing has air conditioning or not.

This dataset contains apartment sale listings scraped from three major real estate platforms in Tunisia:

• Tayara.tn
• Menzili.tn
• Mubawab.tn

Using Scrapy, we collected details such as location, area, number of rooms, floor level, and price. The data is cleaned and preprocessed, making it ideal for:

Predicting Apartment Prices
Real Estate Market Trends Analysis
Machine Learning Model Training

File Information

Features

Why Use This Dataset?

✔️ Real-world real estate data ✔️ Preprocessed & ready for ML models ✔️ Great for price prediction & market analysis

How to Use

Please refer to the src code of models in my GitHub repositpry

https://github.com/Chebil-Ilef/Apartment-Price-Predictor

💡 If you find this dataset useful, please upvote & share!

Tunisia Real Estate Software Market is expected to grow during 2025-2031

Context

Using this dataset you can predict property prices in Tunisia.

Content

I acquired this data by scraping tayara.tn which is a real estate Tunisian website.

In December 2022, the average import price for bearing housings not incorporating ball or roller bearings, plain shaft bearings amounted to $18,658 per ton, with a decrease of -30.8% against the previous month.

In 2024, the Tunisian market for bearing housings not incorporating ball or roller bearings, plain shaft bearings decreased by -1% to $4.4M for the first time since 2019, thus ending a four-year rising trend. Overall, consumption, however, saw a mild decrease. Bearing housing without ball bearing consumption peaked at $5.7M in 2013; however, from 2014 to 2024, consumption failed to regain momentum.

This dataset was scraped from the well-known Tunisian website: Tayara.tn . It includes various data points that can be used for multiple applications, such as natural language processing (NLP) for post verification, price prediction, and more.

Note

As the data was gathered from user posts, it may contain discrepancies or illogical entries. The goal of this dataset is to provide insights into how sellers market their houses and should not be taken as a strict reference.

The Tunisian camping trailer market contracted modestly to $3.9M in 2024, waning by -4.5% against the previous year. The market value increased at an average annual rate of +1.4% from 2012 to 2024; the trend pattern remained relatively stable, with only minor fluctuations being recorded in certain years. Over the period under review, the market reached the peak level at $4.4M in 2014; however, from 2015 to 2024, consumption stood at a somewhat lower figure.