Hey guys. I am currently starting to work on my universtiy project for Fundamentals of Artificial Intelligence class. I would really appreciate if you could suggest me the datasets according to these requirments :<\/p>\n
“Select a dataset that is suitable for a classification task. The student must avoid selecting the Iris dataset or the<\/p>\n
Palmer Archipelago (Antarctica) penguin dataset. In addition, the meaningfulness of the classification has to be<\/p>\n
considered, e.g. it is meaningless to classify continents by the number of Covid-19 cases because, first, there are<\/p>\n
only six continents and new ones will not appear soon, second, the number of Covid-19 cases is not a<\/p>\n
defining characteristic of continents;<\/p>\n
\u2022 it is preferable to select a dataset that is already given in the format of a .csv datafile;<\/p>\n
\u2022 the dataset should be well-documented (there should be information about who created the set, when and what<\/p>\n
the data source is);<\/p>\n
\u2022 the dataset should be of reasonable size (at least 200 data objects);<\/p>\n
\u2022 the dataset should be deeply annotated (there should be information about which features are stored and what<\/p>\n
they mean);<\/p>\n
\u2022 the number of features should be between 5-15;<\/p>\n
\u2022 the dataset should be labelled;<\/p>\n
\u2022 the student must avoid datasets with many Boolean (true\/false, 1\/0, etc.) or categorical type feature (attribute)<\/p>\n
values. It is preferable to use datasets in which most of the attributes are represented by continuous attribute<\/p>\n
values;<\/p>\n
\u2022 you should avoid datasets of unlabelled data (e.g. text corpora and raw images)”<\/p>\n<\/div>\n
submitted by \/u\/kktsrvii <\/a>
[link]<\/a><\/span> [comments]<\/a><\/span><\/p>
