process_mining 0.5.5

Process Mining library for working with (object-centric) event data
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209

use crate::core::{
    event_data::case_centric::utils::activity_projection::EventLogActivityProjection,
    process_models::case_centric::petri_net::petri_net_struct::Transition, PetriNet,
};

use super::full::{alphappp_discover_petri_net, AlphaPPPConfig};
const AUTO_CONFIGS: &[AlphaPPPConfig] = &[
    AlphaPPPConfig {
        balance_thresh: 0.6,
        fitness_thresh: 0.4,
        replay_thresh: 0.0,
        log_repair_skip_df_thresh_rel: 4.0,
        log_repair_loop_df_thresh_rel: 4.0,
        absolute_df_clean_thresh: 1,
        relative_df_clean_thresh: 0.01,
    },
    AlphaPPPConfig {
        balance_thresh: 0.6,
        fitness_thresh: 0.4,
        replay_thresh: 0.0,
        log_repair_skip_df_thresh_rel: 2.0,
        log_repair_loop_df_thresh_rel: 2.0,
        absolute_df_clean_thresh: 1,
        relative_df_clean_thresh: 0.01,
    },
    AlphaPPPConfig {
        balance_thresh: 0.4,
        fitness_thresh: 0.6,
        replay_thresh: 0.0,
        log_repair_skip_df_thresh_rel: 4.0,
        log_repair_loop_df_thresh_rel: 4.0,
        absolute_df_clean_thresh: 1,
        relative_df_clean_thresh: 0.01,
    },
    AlphaPPPConfig {
        balance_thresh: 0.4,
        fitness_thresh: 0.6,
        replay_thresh: 0.0,
        log_repair_skip_df_thresh_rel: 2.0,
        log_repair_loop_df_thresh_rel: 2.0,
        absolute_df_clean_thresh: 1,
        relative_df_clean_thresh: 0.01,
    },
    AlphaPPPConfig {
        balance_thresh: 0.4,
        fitness_thresh: 0.6,
        replay_thresh: 0.0,
        log_repair_skip_df_thresh_rel: 2.0,
        log_repair_loop_df_thresh_rel: 2.0,
        absolute_df_clean_thresh: 5,
        relative_df_clean_thresh: 0.05,
    },
    AlphaPPPConfig {
        balance_thresh: 0.1,
        fitness_thresh: 0.8,
        replay_thresh: 0.0,
        log_repair_skip_df_thresh_rel: 2.0,
        log_repair_loop_df_thresh_rel: 2.0,
        absolute_df_clean_thresh: 5,
        relative_df_clean_thresh: 0.05,
    },
    AlphaPPPConfig {
        balance_thresh: 0.25,
        fitness_thresh: 0.75,
        replay_thresh: 0.0,
        log_repair_skip_df_thresh_rel: 2.0,
        log_repair_loop_df_thresh_rel: 2.0,
        absolute_df_clean_thresh: 25,
        relative_df_clean_thresh: 0.1,
    },
    AlphaPPPConfig {
        balance_thresh: 0.1,
        fitness_thresh: 0.8,
        replay_thresh: 0.0,
        log_repair_skip_df_thresh_rel: 4.0,
        log_repair_loop_df_thresh_rel: 4.0,
        absolute_df_clean_thresh: 1,
        relative_df_clean_thresh: 0.01,
    },
    AlphaPPPConfig {
        balance_thresh: 0.1,
        fitness_thresh: 0.8,
        replay_thresh: 0.0,
        log_repair_skip_df_thresh_rel: 2.0,
        log_repair_loop_df_thresh_rel: 2.0,
        absolute_df_clean_thresh: 1,
        relative_df_clean_thresh: 0.01,
    },
    AlphaPPPConfig {
        balance_thresh: 0.1,
        fitness_thresh: 0.9,
        replay_thresh: 0.0,
        log_repair_skip_df_thresh_rel: 4.0,
        log_repair_loop_df_thresh_rel: 4.0,
        absolute_df_clean_thresh: 1,
        relative_df_clean_thresh: 0.01,
    },
    AlphaPPPConfig {
        balance_thresh: 0.1,
        fitness_thresh: 0.9,
        replay_thresh: 0.0,
        log_repair_skip_df_thresh_rel: 2.0,
        log_repair_loop_df_thresh_rel: 2.0,
        absolute_df_clean_thresh: 1,
        relative_df_clean_thresh: 0.01,
    },
];

/// Automatically select parameters for Alpha+++ and discover a [`PetriNet`] using the chosen parameters
///
/// Currently, tests out multiple paramater configurations and selects a best one based on the discovered Petri net
pub fn alphappp_discover_with_auto_parameters(
    log_proj: &EventLogActivityProjection,
) -> (AlphaPPPConfig, PetriNet) {
    let mut best: Option<(AlphaPPPConfig, f32, PetriNet)> = None;
    for c in AUTO_CONFIGS {
        let pn = alphappp_discover_petri_net(log_proj, *c);
        let score = score_discovered_pn(&pn, c);
        match best {
            Some((_, best_score, _)) => {
                if score > best_score {
                    best = Some((*c, score, pn));
                }
            }
            None => {
                best = Some((*c, score, pn));
            }
        }
    }
    let (best_config, best_score, best_pn) = best.unwrap();
    println!("Best score: {best_score:.2} with config {best_config:?}");
    println!(
        "Resulting net has {} arcs, {} transitions and {} places",
        best_pn.arcs.len(),
        best_pn.transitions.len(),
        best_pn.places.len()
    );
    (best_config, best_pn)
}

fn score_discovered_pn(pn: &PetriNet, config: &AlphaPPPConfig) -> f32 {
    fn is_transition_well_connected(pn: &PetriNet, t: &Transition) -> bool {
        if t.label.is_some() {
            let preset_connected = pn
                .preset_of_transition(t.into())
                .into_iter()
                .filter(|p| {
                    pn.is_in_initial_marking(p)
                        || pn.is_in_a_final_marking(p)
                        || pn
                            .preset_of_place(*p)
                            .into_iter()
                            .filter(|ot_id| {
                                pn.transitions
                                    .get(&ot_id.get_uuid())
                                    .unwrap()
                                    .label
                                    .is_some()
                            })
                            .count()
                            >= 1
                })
                .count()
                >= 1;

            let postset_connected = pn
                .postset_of_transition(t.into())
                .into_iter()
                .filter(|p| {
                    pn.is_in_initial_marking(p)
                        || pn.is_in_a_final_marking(p)
                        || pn
                            .postset_of_place(*p)
                            .into_iter()
                            .filter(|ot_id| {
                                pn.transitions
                                    .get(&ot_id.get_uuid())
                                    .unwrap()
                                    .label
                                    .is_some()
                            })
                            .count()
                            >= 1
                })
                .count()
                >= 1;

            preset_connected && postset_connected
        } else {
            false
        }
    }
    let num_disconnected_trans = pn
        .transitions
        .clone()
        .into_iter()
        .filter(|(_, t)| !is_transition_well_connected(pn, t))
        .count();

    config.fitness_thresh
        * (1.0 - config.balance_thresh)
        * (1.0
            - (num_disconnected_trans as f32
                / pn.transitions
                    .iter()
                    .filter(|(_, t)| t.label.is_some())
                    .count() as f32))
            .powf(2.0)
}