Synthetic data are artificially created data that resemble real world data. It is useful for training on machine learning and deep learning as it is expensive to obtain real world data. Millions of bytes of synthetic data can be generated algorithmically in reasonable time. In certain circumstances, it can be used to fill gaps when real world data is non-existent or time consuming. It can solve problems in different sectors by enriching with data which otherwise not attainable.
Users can download synthetic datasets from this website for free. The word ‘synthetic’ means artificial. That is, the data in these datasets are generated artificially during run time and does not include any real-world data. Users can select the desired columns or all columns which are listed as check boxes and click submit button to obtain the desired dataset. All columns can be selected in one click by checking the ‘Select All’ check box.
Once the desired columns are selected in a dataset and submit button is clicked then the first five or ten rows of desired dataset can be viewed as a table structure. The samples in the dataset generated are not fixed or constant and will be different each time the submit button is clicked. Likewise, numerous sets of datasets with different samples can be generated. The dataset once generated can be downloaded as a .csv file by clicking the ‘Download .csv’ link.
The datasets thus downloaded can be subjected for Data cleaning, Data Analysis and visualisation, model building and creating dashboards.
The mail data can be utilized for Natural Language Processing (NLP) tasks.The dataset comprises 500 rows, each of which identifies whether the mail is spam or not spam.The columns in the dataset are 'Mail' and 'Mail Category', and the mail content is categorized as 'Spam' or 'Not Spam'. The dataset must be cleaned and then subjected to analysis and visualisation. A classification model can be constructed from the dataset.
More than one mail dataset with different set of samples can be generated and can be combined them to get an enlarged dataset.
There is a rise of cardio vascular disease among the population through the years due to poor lifestyle and stressful life. It is necessary for an earlier management of the disease.
The cardio vascular disease dataset that can be downloaded consists of 500 rows each of which shows the risk factors of the disease and based on this the presence of the disease. The columns in the cardio vascular disease data are ‘Gender’, ‘Age’, ‘Smoking’, ‘BP’, ‘Cholesterol’, ‘Diabetes’, ‘Weight’, ‘Alcohol consumption’, ‘Heart Disease Risk’.
A prediction model can be built from this dataset to predict heart disease risk. Two datasets can be downloaded for this of which one can be used as train and other can be used as test.
The sales dataset is a time series one. It contains information about sales in different regions over a particular period and contains 500 rows. The columns in the sales data are ‘OrderID’, ‘CustomerID’, ‘CustomerName’, ‘Region’, ‘ProductID’, ‘ProductName’, ‘OrderDate’, ‘Category’, ‘SubCategory’, ‘Quantity’, ‘Discount’, ‘Price’, ‘TotalAmount’. Analytical Dashboards can be built from the data.