Python machine learning model for binary classification

A Python machine learning model is needed for binary classification.

The raw data comes in two sets. One set of positive samples ("Good basket"), one set of negative ones ("Bad basket").

Good basket:
['Mangosteen', 'Cranberry', 'Tamarillo', 'Loquat', 'Mango']
['Lemon', 'Yuzu', 'Strawberry', 'Lulo', 'Ugli_fruit']
['Coconut', 'Kiwifruit', 'Ximenia', 'Mage_berry', 'Coco_de_mer']
['Melon', 'Longan', 'Date', 'Marionberry', 'Tayberry']
['Cherry', 'Mangosteen', 'Crab_apple', 'Mamey_Sapote', 'Tamarind']

Bad basket:
['Star_apple', 'Melon', 'Loquat', 'Currant', 'Lulo']
['White_sapote', 'Lime', 'Mango', 'Marionberry', 'Lulo']
['Mamey_Apple', 'Strawberry', 'Mangosteen', 'Mage_berry', 'Loquat']
['Lychee', 'Yuzu', 'Damson', 'Ximenia', 'Star_fruit']
['Ximenia', 'Longan', 'Star_fruit', 'Mangosteen', 'Ugli_fruit']

The format of data set comes in the form of list. The small challenge here is - the variables are in the data, as samples below.
(The actual list elements are in Asian (Chinese) characters, which is not a barrier.)

Note:
1. No restriction on which machine learning methold/algorithm to use.
2. The prediction accuracy doesn't need to be perfect.
3. A final selected model with information on accuracy of prediction.
4. It is to be completed in 10 days since the project is awarded.

Wanted result:
When a line of new data put into the trained model, it tells if the new line is likely to be positive or negative.