Intro

Why temporal - cite all articles, all our references

usual part 2 → give background about our work - entire pipeline

outliers what to do, utmost k loss, the threshold on k, how we are considering distribution, priority,

examples

evaluate on the metric we made, all metrics which are applicable,

Results

The equation is fixed

For each attribute it is the score / probability of finding it+1 → to avoid / by 0 error

The total score of each tuple is the size of domain for categoric attributes * score while for numeric there is no such consideration

in the following x, y coordinates, x is the k value can go from 50...500 and y is the cost of degradation calculated overall

import matplotlib.pyplot as plt

import numpy as np

x = [50, 100, 150, 200, 250, 300, 350, 400, 450, 500]

y = [3.0309240974654497, 3.9527959480037045, 5.134053790613963, 5.826214512015462, 6.570773455027218, 8.13954910601862, 8.781923793489744, 9.238902650923666, 11.111952525363902, 11.037898086441698]