after applying levenshtein distance algorithm I get a dataframe like this:

Elemento_lista column is the index of the element that is compared to others, Item_ID is the id of the element, Score is Score generated by the algorithm, idx is the index of the element that was found as similar (same as Elemento_lista , but for elements that were found as similar), ITEM_ID_Coincidencia is the id of the element found as similar

It´s a small sample of the real DF (More than 300000 rows), I´ll need to drop lines that are the same , for example...if Elemento_lista 4, is equal to idx 5,6,and 7...they are all the same, so I don't need lines where 5 is equal to 4, 6 and 7/ 6 is equal to 4,5,7 and 7 is equal to 4,5,6. The same for each Elemento_Lista : value=9 is equal to idx 10, so...I don't need the line Elemento_Lista 10 is equal to idx 9...How could I drop these lines in order to reduce DF len ???

Final DF should be:

I don´t know how to do this...is it possible?

Thanks in advance

CodePudding user response：

Preparing data like example:

a = [
[4,691776,100,5,691777],
[4,691776,100,6,691789],
[4,691776,100,7,691791],
[5,691777,100,4,691776],
[5,691777,100,6,691789],
[5,691777,100,7,691791],
[6,691789,100,4,691776],
[6,691789,100,5,691777],
[6,691789,100,7,691791],
[7,691791,100,4,691776],
[7,691791,100,5,691777],
[7,691791,100,6,691789],
[9,1407402,100,10,1407424],
[10,1407424,100,9,1407402]
]
c = ['Elemento_lista', 'Item_ID', 'Score', 'idx', 'ITEM_ID_Coincidencia']
df = pd.DataFrame(data = a, columns = c)
df

Now, you insert one column: it will contain an array of 2 sorted indexes.

tuples_of_indexes = [sorted([x[0], x[3]]) for x in df.values]
df.insert(5, 'tuple_of_indexes', (tuples_of_indexes))

Then all the dataframe is sorted by the inserted column:

df = df.sort_values(by=['tuple_of_indexes'])

Then you eliminate rows that repeat inserted column:

df = df[~df['tuple_of_indexes'].apply(tuple).duplicated()]

For last, you eliminate inserted column: 'tuple_of_indexes':

df.drop(['tuple_of_indexes'], axis=1)

The output is:

Elemento_lista  Item_ID Score   idx ITEM_ID_Coincidencia
0   4   691776  100 5   691777
1   4   691776  100 6   691789
2   4   691776  100 7   691791
4   5   691777  100 6   691789
5   5   691777  100 7   691791
8   6   691789  100 7   691791
12  9   1407402 100 10  1407424

CodePudding user response：

This can be approached using graph theory.

You have the following relationships between your IDs:

So what you need to do is find the subgraphs.

For this we can use

You can change strategy here and use a directed graph and