Initial version

.gitignore

@@ -0,0 +1 @@
+.vscode/

app.py

@@ -0,0 +1,106 @@
+import streamlit as st
+import pandas as pd
+
+import plotly.graph_objects as go
+import plotly.express as px
+
+from ast import literal_eval
+from lxml import etree as ET
+
+
+def prepare_data():
+    data = pd.read_csv(
+        "data/2002829_mapped_roles.csv", index_col=0, 
+        converters={"frame": literal_eval, "changed_roles": literal_eval, "unchanged_roles": literal_eval, "roles": literal_eval}
+    )
+    
+    frame_to_scenario, frame_to_super_scenario = load_kicktionary_info()
+
+    # extract information from "frame" tuples, filter & reorder columns
+    data_ = (
+        data
+            .assign(sentence_idx=data["frame"].apply(lambda frame: frame[0]))
+            .assign(frame_idx=data["frame"].apply(lambda frame: frame[1]))
+            .assign(frame_name=data["frame"].apply(lambda frame: frame[2]))
+            .assign(frame_scenario=data["frame"].apply(lambda frame: frame_to_scenario[frame[2]]))
+            .assign(frame_super_scenario=data["frame"].apply(lambda frame: frame_to_super_scenario[frame[2]]))
+            .assign(frame_target=data["frame"].apply(lambda frame: frame[3]))
+            .drop(columns=["frame"])
+    )[["sentence_idx", "frame_idx", "frame_name", "frame_scenario", "frame_super_scenario", "frame_target", "changed_roles", "roles"]]
+
+    # assign value in 0 < t < 1 to represent each frame instance's "time" point in the article
+    max_sent = max(data_["sentence_idx"])
+    max_frame_per_sent = data_.groupby("sentence_idx").agg({"frame_idx": max}).reset_index()
+    sent_to_max_frame = dict(zip(max_frame_per_sent["sentence_idx"], max_frame_per_sent["frame_idx"]))
+    data_with_time = data_.assign(
+        time_point= (data_
+            .apply(lambda row: (row["sentence_idx"] + row["frame_idx"] / (sent_to_max_frame[row["sentence_idx"]])) / (max_sent + 1), axis=1)
+        )
+    )
+    data_with_first_roles = data_with_time.assign(
+        first_role = data_with_time["changed_roles"].apply(lambda roles: roles[0] if len(roles) > 0 else None)
+    )
+    return data_with_first_roles
+
+
+def load_kicktionary_info():
+
+    kicktionary = ET.parse("kicktionary_lu_info.xml")
+    frame_to_scenario = {
+        lu.attrib["frame"]: lu.attrib["scenario"]
+        for lu in kicktionary.xpath(".//LEXICAL-UNIT") if lu.attrib["frame"]
+    }
+    frame_to_super_scenario = {
+        lu.attrib["frame"]: lu.attrib["super-scenario"]
+        for lu in kicktionary.xpath(".//LEXICAL-UNIT") if lu.attrib["frame"]
+    }
+
+    return frame_to_scenario, frame_to_super_scenario
+
+
+def explore_timeline():
+
+    with st.container():
+
+        st.title("Football Perspective Chains")
+        
+        frame_label_map = {
+            "frame_name": "frames",
+            "frame_scenario": "scenarios (groups of related frames)",
+            "frame_super_scenario": "super scenarios (groups of related scenarios)"
+        }
+        frame_column = st.selectbox(
+            label="Display frames as: ", 
+            options=("frame_name", "frame_scenario", "frame_super_scenario"),
+            format_func=lambda label: frame_label_map[label]
+        )
+
+        st.header("Timeline")
+        data = prepare_data()
+        time_scatter = (
+            data
+                .dropna(axis=0, subset=["first_role"])
+                .plot.scatter(
+                    x="first_role", y="time_point", backend="plotly", color=frame_column
+                )
+        )
+        time_scatter.update_traces(marker_size=20)
+        time_scatter.update_layout(height=1000)
+        st.plotly_chart(time_scatter)
+
+        st.header("Overall focus")
+        focus_bar = data.dropna(axis=0, subset=["first_role"])["first_role"].value_counts().plot.bar(y="first_role", backend="plotly")
+        st.plotly_chart(focus_bar)
+
+        st.header("Focus by frame")
+
+        for team in ["Man. United", "Rangers"]:
+            st.subheader(team)
+            frame_bar = data.dropna(axis=0, subset=["first_role"])["frame_scenario"].value_counts().plot.bar(y=frame_column, backend="plotly")
+            st.plotly_chart(frame_bar)
+
+
+
+
+if __name__ == "__main__":
+    explore_timeline()

data/2002829_mapped_roles.csv

@@ -0,0 +1,40 @@
+,frame,changed_roles,unchanged_roles,roles,changed_role
+0,"(0, 0, 'Draw', 'draw', 'NOUN')",['Rangers'],"[('COMPETITION_STAGE', ['from', 'a', 'tight', 'Anglo-Scottish', 'Group', 'C', 'contest']), ('FINAL_SCORE', ['goalless']), ('MATCH_LOCATION', ['at', 'Old', 'Trafford'])]","[('COMPETITION_STAGE', ['from', 'a', 'tight', 'Anglo-Scottish', 'Group', 'C', 'contest']), ('FINAL_SCORE', ['goalless']), ('MATCH_LOCATION', ['at', 'Old', 'Trafford']), ('TEAM1', ['Rangers', 'FC'])]",Changed Something
+1,"(0, 1, 'Match', 'contest', 'NOUN')",[],"[('MATCH_LOCATION', ['at', 'Old', 'Trafford']), ('COMPETITION_STAGE', ['Anglo-Scottish', 'Group', 'C'])]","[('MATCH_LOCATION', ['at', 'Old', 'Trafford']), ('COMPETITION_STAGE', ['Anglo-Scottish', 'Group', 'C'])]",No Changed
+2,"(1, 0, 'Competition', 'titleholders', 'NOUN')",[],"[('TEAM', ['Scottish'])]","[('TEAM', ['Scottish'])]",No Changed
+3,"(1, 1, 'Concede_Goal', 'conceded', 'VERB')",['Rangers'],"[('CONCEDING_TEAM', ['they']), ('GOAL', ['more', 'goals'])]","[('CONCEDING_TEAM', ['they']), ('CONCEDING_TEAM', ['Walter', 'Smith', ""'s"", 'Scottish', 'titleholders']), ('GOAL', ['more', 'goals'])]",Changed Something
+4,"(1, 2, 'Competition_Stage', 'group', 'NOUN')",[],"[('COMPETITION', ['UEFA', 'Champions', 'League'])]","[('COMPETITION', ['UEFA', 'Champions', 'League'])]",No Changed
+5,"(1, 3, 'Bring_Off', 'return', 'VERB')",['Man. United'],"[('TARGET', ['north', 'of', 'the', 'border'])]","[('SUBSTITUTED_PLAYER', ['Antonio', 'Valencia']), ('TARGET', ['north', 'of', 'the', 'border'])]",Changed Something
+6,"(1, 6, 'Bring_Off', 'departing', 'VERB')",['Man. United'],[],"[('SUBSTITUTED_PLAYER', ['Antonio', 'Valencia'])]",All changed
+7,"(1, 8, 'Match_Temporal_Subdivision', 'period', 'NOUN')",[],"[('FIRST_OR_SECOND', ['second'])]","[('FIRST_OR_SECOND', ['second'])]",No Changed
+8,"(2, 0, 'Team', 'team', 'NOUN')",[],"[('TEAM', ['Scottish'])]","[('TEAM', ['Scottish'])]",No Changed
+9,"(2, 1, 'Competition', 'elite', 'NOUN')",[],"[('COMPETITION', ['Europe', ""'s""])]","[('COMPETITION', ['Europe', ""'s""])]",No Changed
+10,"(2, 2, 'Competition', 'competition', 'NOUN')",[],"[('COMPETITION', ['Europe', ""'s""])]","[('COMPETITION', ['Europe', ""'s""])]",No Changed
+11,"(2, 3, 'Victory', 'triumph', 'NOUN')",['Man. United'],[],"[('MATCH_LOCATION', ['at', 'Leeds', 'United', 'AFC'])]",All changed
+12,"(2, 4, 'Deploy', 'fielding', 'NOUN')",['Rangers'],"[('COACH', ['his']), ('PLAYER', ['of', 'three', 'central', 'defenders'])]","[('COACH', ['his']), ('PLAYER', ['of', 'three', 'central', 'defenders']), ('PLAYER', ['in', 'Madjid', 'Bougherra', ',', 'David', 'Weir', 'and', 'Saša', 'Papac'])]",Changed Something
+13,"(2, 5, 'Player', 'full-backs', 'NOUN')",['Rangers'],[],"[('PLAYER', ['Steven', 'Whittaker', 'and', 'Kirk', 'Broadfoot'])]",All changed
+14,"(3, 1, 'Draw', 'draw', 'NOUN')",['Man. United'],"[('MATCH_LOCATION', ['at', 'Everton'])]","[('TEAM1', ['United', ""'s""]), ('MATCH_LOCATION', ['at', 'Everton'])]",Changed Something
+15,"(3, 2, 'Shot', 'nod', 'VERB')",['Man. United'],"[('SHOOTER', ['he']), ('TARGET', ['just', 'wide', 'of', 'the', 'near', 'post'])]","[('MOVING_BALL', ['Fabio', ""'s"", 'right-wing', 'cross']), ('SHOOTER', ['he']), ('TARGET', ['just', 'wide', 'of', 'the', 'near', 'post'])]",Changed Something
+16,"(3, 3, 'Pass', 'cross', 'NOUN')",['Man. United'],"[('TARGET', ['right-wing'])]","[('TARGET', ['right-wing']), ('PASSER', ['Fabio', ""'s""])]",Changed Something
+17,"(3, 4, 'Goal_Target', 'post', 'NOUN')",[],"[('DISTANCE_TO_BALL', ['near'])]","[('DISTANCE_TO_BALL', ['near'])]",No Changed
+18,"(3, 5, 'One_On_One', 'two-on-one', 'NOUN')","['Man. United', 'Rangers', 'Man. United']","[('PLAYER1', ['he'])]","[('MOVING_BALL', ['Fabio', ""'s"", 'right-wing', 'cross']), ('PLAYER2', ['with', 'Bougherra']), ('PLAYER1', ['he']), ('PLAYER1', ['he', 'and', 'the', 'recalled', 'Wayne', 'Rooney'])]",Changed Something
+19,"(3, 6, 'Pass', 'pass', 'NOUN')",[],"[('PASSER', ['his'])]","[('PASSER', ['his'])]",No Changed
+20,"(3, 7, 'Intervene', 'avert', 'VERB')",['Rangers'],"[('SHOT', ['the', 'threat'])]","[('SHOT', ['the', 'threat']), ('INTERVENING_PLAYER', ['Weir'])]",Changed Something
+21,"(4, 0, 'Shot', 'drive', 'NOUN')",[],"[('PATH', ['low'])]","[('PATH', ['low'])]",No Changed
+22,"(4, 1, 'Ball_Move', 'whistled', 'VERB')",[],"[('MOVING_BALL', ['a', 'low', 'drive']), ('TARGET', ['past', 'the', 'goal', 'frame'])]","[('MOVING_BALL', ['a', 'low', 'drive']), ('TARGET', ['past', 'the', 'goal', 'frame'])]",No Changed
+23,"(4, 3, 'Player', 'midfielder', 'NOUN')",[],"[('TEAM', ['Irish'])]","[('TEAM', ['Irish'])]",No Changed
+24,"(4, 5, 'Shot', 'volleying', 'VERB')",[],"[('SOURCE', ['from', 'the', 'edge', 'of', 'the', 'area']), ('TARGET', ['wide']), ('SHOOTER', ['the', 'Irish', 'midfielder'])]","[('SOURCE', ['from', 'the', 'edge', 'of', 'the', 'area']), ('TARGET', ['wide']), ('SHOOTER', ['the', 'Irish', 'midfielder'])]",No Changed
+25,"(4, 7, 'Being_Free', 'isolated', 'VERB')",['Rangers'],[],"[('RECIPIENT', ['Kenny', 'Miller'])]",All changed
+26,"(4, 9, 'Team', 'defence', 'NOUN')",['Man. United'],[],"[('TEAM', ['United', ""'s""])]",All changed
+27,"(5, 0, 'Control', 'got', 'VERB')","['Rangers', 'Man. United', 'Rangers']",[],"[('RECIPIENT', ['Kirk', 'Broadfoot']), ('RECIPIENT', ['Darren', 'Fletcher']), ('RECIPIENT', ['Papac'])]",All changed
+28,"(5, 1, 'Pass', 'cross', 'NOUN')",['Rangers'],[],"[('PASSER', ['Steven', 'Naismith'])]",All changed
+29,"(5, 2, 'Intercept', 'interception', 'NOUN')",[],"[('INTERCEPTOR', ['his', 'own'])]","[('INTERCEPTOR', ['his', 'own'])]",No Changed
+30,"(5, 3, 'Challenge', 'challenge', 'NOUN')",['Man. United'],[],"[('OPPONENT_PLAYER', ['Smalling'])]",All changed
+31,"(5, 4, 'Field', 'edge', 'NOUN')",['Man. United'],[],"[('TEAM', ['United'])]",All changed
+32,"(5, 5, 'Referee', 'referee', 'VERB')",[],"[('REFEREE', ['Olegário', 'Benquerença'])]","[('REFEREE', ['Olegário', 'Benquerença'])]",No Changed
+33,"(5, 6, 'Control', 'latching', 'VERB')",['Rangers'],[],"[('RECIPIENT', ['Kirk', 'Broadfoot'])]",All changed
+34,"(5, 7, 'Shot', 'lofted', 'VERB')",['Man. United'],"[('TARGET', ['over', 'the', 'top'])]","[('TARGET', ['over', 'the', 'top']), ('SHOOTER', ['Ryan', 'Giggs', ""'s""])]",Changed Something
+35,"(5, 8, 'Pass', 'ball', 'NOUN')",['Man. United'],"[('TARGET', ['over', 'the', 'top'])]","[('TARGET', ['over', 'the', 'top']), ('PASSER', ['Ryan', 'Giggs', ""'s""])]",Changed Something
+36,"(5, 10, 'Shot_Supports', 'shot', 'VERB')",['Man. United'],[],"[('SHOOTER', ['Gibson'])]",All changed
+37,"(5, 11, 'Shot', 'whisker', 'NOUN')",['Man. United'],"[('TARGET', ['over'])]","[('TARGET', ['over']), ('SHOOTER', ['Gibson'])]",Changed Something
+38,"(5, 13, 'Score_Goal', 'claim', 'VERB')",[],"[('GOAL', ['their', 'point']), ('REFEREE', ['referee', 'Olegário', 'Benquerença'])]","[('GOAL', ['their', 'point']), ('REFEREE', ['referee', 'Olegário', 'Benquerença'])]",No Changed

data/2002829_temporal_data.json

@@ -0,0 +1,29 @@
+{
+    "Rangers": [
+        "(0, 0, 'Draw', 'draw', 'NOUN')",
+        "(1, 1, 'Concede_Goal', 'conceded', 'VERB')",
+        "(2, 4, 'Deploy', 'fielding', 'NOUN')",
+        "(2, 5, 'Player', 'full-backs', 'NOUN')",
+        "(3, 7, 'Intervene', 'avert', 'VERB')",
+        "(4, 7, 'Being_Free', 'isolated', 'VERB')",
+        "(5, 0, 'Control', 'got', 'VERB')",
+        "(5, 1, 'Pass', 'cross', 'NOUN')",
+        "(5, 6, 'Control', 'latching', 'VERB')"
+    ],
+    "Man. United": [
+        "(1, 3, 'Bring_Off', 'return', 'VERB')",
+        "(1, 6, 'Bring_Off', 'departing', 'VERB')",
+        "(2, 3, 'Victory', 'triumph', 'NOUN')",
+        "(3, 1, 'Draw', 'draw', 'NOUN')",
+        "(3, 2, 'Shot', 'nod', 'VERB')",
+        "(3, 3, 'Pass', 'cross', 'NOUN')",
+        "(3, 5, 'One_On_One', 'two-on-one', 'NOUN')",
+        "(4, 9, 'Team', 'defence', 'NOUN')",
+        "(5, 3, 'Challenge', 'challenge', 'NOUN')",
+        "(5, 4, 'Field', 'edge', 'NOUN')",
+        "(5, 7, 'Shot', 'lofted', 'VERB')",
+        "(5, 8, 'Pass', 'ball', 'NOUN')",
+        "(5, 10, 'Shot_Supports', 'shot', 'VERB')",
+        "(5, 11, 'Shot', 'whisker', 'NOUN')"
+    ]
+}

notebook.ipynb

@@ -0,0 +1,852 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import os\n",
+    "from rich import print\n",
+    "os.chdir(\"/home/p289731/NWO_FrameNet/Code/football-chains-viz\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from ast import literal_eval"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">(</span><span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">39</span>, <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">5</span><span style=\"font-weight: bold\">)</span>\n",
+       "</pre>\n"
+      ],
+      "text/plain": [
+       "\u001b[1m(\u001b[0m\u001b[1;36m39\u001b[0m, \u001b[1;36m5\u001b[0m\u001b[1m)\u001b[0m\n"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>frame</th>\n",
+       "      <th>changed_roles</th>\n",
+       "      <th>unchanged_roles</th>\n",
+       "      <th>roles</th>\n",
+       "      <th>changed_role</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>(0, 0, Draw, draw, NOUN)</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(COMPETITION_STAGE, [from, a, tight, Anglo-Sc...</td>\n",
+       "      <td>[(COMPETITION_STAGE, [from, a, tight, Anglo-Sc...</td>\n",
+       "      <td>Changed Something</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>(0, 1, Match, contest, NOUN)</td>\n",
+       "      <td>[]</td>\n",
+       "      <td>[(MATCH_LOCATION, [at, Old, Trafford]), (COMPE...</td>\n",
+       "      <td>[(MATCH_LOCATION, [at, Old, Trafford]), (COMPE...</td>\n",
+       "      <td>No Changed</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>(1, 0, Competition, titleholders, NOUN)</td>\n",
+       "      <td>[]</td>\n",
+       "      <td>[(TEAM, [Scottish])]</td>\n",
+       "      <td>[(TEAM, [Scottish])]</td>\n",
+       "      <td>No Changed</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>(1, 1, Concede_Goal, conceded, VERB)</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(CONCEDING_TEAM, [they]), (GOAL, [more, goals])]</td>\n",
+       "      <td>[(CONCEDING_TEAM, [they]), (CONCEDING_TEAM, [W...</td>\n",
+       "      <td>Changed Something</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>(1, 2, Competition_Stage, group, NOUN)</td>\n",
+       "      <td>[]</td>\n",
+       "      <td>[(COMPETITION, [UEFA, Champions, League])]</td>\n",
+       "      <td>[(COMPETITION, [UEFA, Champions, League])]</td>\n",
+       "      <td>No Changed</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                     frame changed_roles  \\\n",
+       "0                 (0, 0, Draw, draw, NOUN)     [Rangers]   \n",
+       "1             (0, 1, Match, contest, NOUN)            []   \n",
+       "2  (1, 0, Competition, titleholders, NOUN)            []   \n",
+       "3     (1, 1, Concede_Goal, conceded, VERB)     [Rangers]   \n",
+       "4   (1, 2, Competition_Stage, group, NOUN)            []   \n",
+       "\n",
+       "                                     unchanged_roles  \\\n",
+       "0  [(COMPETITION_STAGE, [from, a, tight, Anglo-Sc...   \n",
+       "1  [(MATCH_LOCATION, [at, Old, Trafford]), (COMPE...   \n",
+       "2                               [(TEAM, [Scottish])]   \n",
+       "3  [(CONCEDING_TEAM, [they]), (GOAL, [more, goals])]   \n",
+       "4         [(COMPETITION, [UEFA, Champions, League])]   \n",
+       "\n",
+       "                                               roles       changed_role  \n",
+       "0  [(COMPETITION_STAGE, [from, a, tight, Anglo-Sc...  Changed Something  \n",
+       "1  [(MATCH_LOCATION, [at, Old, Trafford]), (COMPE...         No Changed  \n",
+       "2                               [(TEAM, [Scottish])]         No Changed  \n",
+       "3  [(CONCEDING_TEAM, [they]), (CONCEDING_TEAM, [W...  Changed Something  \n",
+       "4         [(COMPETITION, [UEFA, Champions, League])]         No Changed  "
+      ]
+     },
+     "execution_count": 21,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data = pd.read_csv(\n",
+    "    \"data/2002829_mapped_roles.csv\", index_col=0, \n",
+    "    converters={\"frame\": literal_eval, \"changed_roles\": literal_eval, \"unchanged_roles\": literal_eval, \"roles\": literal_eval}\n",
+    "    )\n",
+    "print(data.shape)\n",
+    "data.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>sentence_idx</th>\n",
+       "      <th>frame_idx</th>\n",
+       "      <th>frame_name</th>\n",
+       "      <th>frame_target</th>\n",
+       "      <th>changed_roles</th>\n",
+       "      <th>roles</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>Draw</td>\n",
+       "      <td>draw</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(COMPETITION_STAGE, [from, a, tight, Anglo-Sc...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>Match</td>\n",
+       "      <td>contest</td>\n",
+       "      <td>[]</td>\n",
+       "      <td>[(MATCH_LOCATION, [at, Old, Trafford]), (COMPE...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>Competition</td>\n",
+       "      <td>titleholders</td>\n",
+       "      <td>[]</td>\n",
+       "      <td>[(TEAM, [Scottish])]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>Concede_Goal</td>\n",
+       "      <td>conceded</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(CONCEDING_TEAM, [they]), (CONCEDING_TEAM, [W...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>1</td>\n",
+       "      <td>2</td>\n",
+       "      <td>Competition_Stage</td>\n",
+       "      <td>group</td>\n",
+       "      <td>[]</td>\n",
+       "      <td>[(COMPETITION, [UEFA, Champions, League])]</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   sentence_idx  frame_idx         frame_name  frame_target changed_roles  \\\n",
+       "0             0          0               Draw          draw     [Rangers]   \n",
+       "1             0          1              Match       contest            []   \n",
+       "2             1          0        Competition  titleholders            []   \n",
+       "3             1          1       Concede_Goal      conceded     [Rangers]   \n",
+       "4             1          2  Competition_Stage         group            []   \n",
+       "\n",
+       "                                               roles  \n",
+       "0  [(COMPETITION_STAGE, [from, a, tight, Anglo-Sc...  \n",
+       "1  [(MATCH_LOCATION, [at, Old, Trafford]), (COMPE...  \n",
+       "2                               [(TEAM, [Scottish])]  \n",
+       "3  [(CONCEDING_TEAM, [they]), (CONCEDING_TEAM, [W...  \n",
+       "4         [(COMPETITION, [UEFA, Champions, League])]  "
+      ]
+     },
+     "execution_count": 30,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# max_roles_per_frame = \n",
+    "data_ = (\n",
+    "    data\n",
+    "        .assign(sentence_idx=data[\"frame\"].apply(lambda frame: frame[0]))\n",
+    "        .assign(frame_idx=data[\"frame\"].apply(lambda frame: frame[1]))\n",
+    "        .assign(frame_name=data[\"frame\"].apply(lambda frame: frame[2]))\n",
+    "        .assign(frame_target=data[\"frame\"].apply(lambda frame: frame[3]))\n",
+    "        .drop(columns=[\"frame\"])\n",
+    ")[[\"sentence_idx\", \"frame_idx\", \"frame_name\", \"frame_target\", \"changed_roles\", \"roles\"]]\n",
+    "data_.head()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "max_sent = max(data_[\"sentence_idx\"])\n",
+    "max_frame_per_sent = data_.groupby(\"sentence_idx\").agg({\"frame_idx\": max}).reset_index()\n",
+    "sent_to_max_frame = dict(zip(max_frame_per_sent[\"sentence_idx\"], max_frame_per_sent[\"frame_idx\"]))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 64,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>sentence_idx</th>\n",
+       "      <th>frame_idx</th>\n",
+       "      <th>frame_name</th>\n",
+       "      <th>frame_target</th>\n",
+       "      <th>changed_roles</th>\n",
+       "      <th>roles</th>\n",
+       "      <th>time_point</th>\n",
+       "      <th>first_role</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>Draw</td>\n",
+       "      <td>draw</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(COMPETITION_STAGE, [from, a, tight, Anglo-Sc...</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>Rangers</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>Concede_Goal</td>\n",
+       "      <td>conceded</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(CONCEDING_TEAM, [they]), (CONCEDING_TEAM, [W...</td>\n",
+       "      <td>0.187500</td>\n",
+       "      <td>Rangers</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>1</td>\n",
+       "      <td>3</td>\n",
+       "      <td>Bring_Off</td>\n",
+       "      <td>return</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(SUBSTITUTED_PLAYER, [Antonio, Valencia]), (T...</td>\n",
+       "      <td>0.229167</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>1</td>\n",
+       "      <td>6</td>\n",
+       "      <td>Bring_Off</td>\n",
+       "      <td>departing</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(SUBSTITUTED_PLAYER, [Antonio, Valencia])]</td>\n",
+       "      <td>0.291667</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td>2</td>\n",
+       "      <td>3</td>\n",
+       "      <td>Victory</td>\n",
+       "      <td>triumph</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(MATCH_LOCATION, [at, Leeds, United, AFC])]</td>\n",
+       "      <td>0.433333</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td>2</td>\n",
+       "      <td>4</td>\n",
+       "      <td>Deploy</td>\n",
+       "      <td>fielding</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(COACH, [his]), (PLAYER, [of, three, central,...</td>\n",
+       "      <td>0.466667</td>\n",
+       "      <td>Rangers</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td>2</td>\n",
+       "      <td>5</td>\n",
+       "      <td>Player</td>\n",
+       "      <td>full-backs</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(PLAYER, [Steven, Whittaker, and, Kirk, Broad...</td>\n",
+       "      <td>0.500000</td>\n",
+       "      <td>Rangers</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td>3</td>\n",
+       "      <td>1</td>\n",
+       "      <td>Draw</td>\n",
+       "      <td>draw</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(TEAM1, [United, 's]), (MATCH_LOCATION, [at, ...</td>\n",
+       "      <td>0.523810</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>3</td>\n",
+       "      <td>2</td>\n",
+       "      <td>Shot</td>\n",
+       "      <td>nod</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(MOVING_BALL, [Fabio, 's, right-wing, cross])...</td>\n",
+       "      <td>0.547619</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>16</th>\n",
+       "      <td>3</td>\n",
+       "      <td>3</td>\n",
+       "      <td>Pass</td>\n",
+       "      <td>cross</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(TARGET, [right-wing]), (PASSER, [Fabio, 's])]</td>\n",
+       "      <td>0.571429</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>18</th>\n",
+       "      <td>3</td>\n",
+       "      <td>5</td>\n",
+       "      <td>One_On_One</td>\n",
+       "      <td>two-on-one</td>\n",
+       "      <td>[Man. United, Rangers, Man. United]</td>\n",
+       "      <td>[(MOVING_BALL, [Fabio, 's, right-wing, cross])...</td>\n",
+       "      <td>0.619048</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>20</th>\n",
+       "      <td>3</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Intervene</td>\n",
+       "      <td>avert</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(SHOT, [the, threat]), (INTERVENING_PLAYER, [...</td>\n",
+       "      <td>0.666667</td>\n",
+       "      <td>Rangers</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>25</th>\n",
+       "      <td>4</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Being_Free</td>\n",
+       "      <td>isolated</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(RECIPIENT, [Kenny, Miller])]</td>\n",
+       "      <td>0.796296</td>\n",
+       "      <td>Rangers</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>26</th>\n",
+       "      <td>4</td>\n",
+       "      <td>9</td>\n",
+       "      <td>Team</td>\n",
+       "      <td>defence</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(TEAM, [United, 's])]</td>\n",
+       "      <td>0.833333</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>27</th>\n",
+       "      <td>5</td>\n",
+       "      <td>0</td>\n",
+       "      <td>Control</td>\n",
+       "      <td>got</td>\n",
+       "      <td>[Rangers, Man. United, Rangers]</td>\n",
+       "      <td>[(RECIPIENT, [Kirk, Broadfoot]), (RECIPIENT, [...</td>\n",
+       "      <td>0.833333</td>\n",
+       "      <td>Rangers</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>28</th>\n",
+       "      <td>5</td>\n",
+       "      <td>1</td>\n",
+       "      <td>Pass</td>\n",
+       "      <td>cross</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(PASSER, [Steven, Naismith])]</td>\n",
+       "      <td>0.846154</td>\n",
+       "      <td>Rangers</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>30</th>\n",
+       "      <td>5</td>\n",
+       "      <td>3</td>\n",
+       "      <td>Challenge</td>\n",
+       "      <td>challenge</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(OPPONENT_PLAYER, [Smalling])]</td>\n",
+       "      <td>0.871795</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>31</th>\n",
+       "      <td>5</td>\n",
+       "      <td>4</td>\n",
+       "      <td>Field</td>\n",
+       "      <td>edge</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(TEAM, [United])]</td>\n",
+       "      <td>0.884615</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>33</th>\n",
+       "      <td>5</td>\n",
+       "      <td>6</td>\n",
+       "      <td>Control</td>\n",
+       "      <td>latching</td>\n",
+       "      <td>[Rangers]</td>\n",
+       "      <td>[(RECIPIENT, [Kirk, Broadfoot])]</td>\n",
+       "      <td>0.910256</td>\n",
+       "      <td>Rangers</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>34</th>\n",
+       "      <td>5</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Shot</td>\n",
+       "      <td>lofted</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(TARGET, [over, the, top]), (SHOOTER, [Ryan, ...</td>\n",
+       "      <td>0.923077</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>35</th>\n",
+       "      <td>5</td>\n",
+       "      <td>8</td>\n",
+       "      <td>Pass</td>\n",
+       "      <td>ball</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(TARGET, [over, the, top]), (PASSER, [Ryan, G...</td>\n",
+       "      <td>0.935897</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>36</th>\n",
+       "      <td>5</td>\n",
+       "      <td>10</td>\n",
+       "      <td>Shot_Supports</td>\n",
+       "      <td>shot</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(SHOOTER, [Gibson])]</td>\n",
+       "      <td>0.961538</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>37</th>\n",
+       "      <td>5</td>\n",
+       "      <td>11</td>\n",
+       "      <td>Shot</td>\n",
+       "      <td>whisker</td>\n",
+       "      <td>[Man. United]</td>\n",
+       "      <td>[(TARGET, [over]), (SHOOTER, [Gibson])]</td>\n",
+       "      <td>0.974359</td>\n",
+       "      <td>Man. United</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "    sentence_idx  frame_idx     frame_name frame_target  \\\n",
+       "0              0          0           Draw         draw   \n",
+       "3              1          1   Concede_Goal     conceded   \n",
+       "5              1          3      Bring_Off       return   \n",
+       "6              1          6      Bring_Off    departing   \n",
+       "11             2          3        Victory      triumph   \n",
+       "12             2          4         Deploy     fielding   \n",
+       "13             2          5         Player   full-backs   \n",
+       "14             3          1           Draw         draw   \n",
+       "15             3          2           Shot          nod   \n",
+       "16             3          3           Pass        cross   \n",
+       "18             3          5     One_On_One   two-on-one   \n",
+       "20             3          7      Intervene        avert   \n",
+       "25             4          7     Being_Free     isolated   \n",
+       "26             4          9           Team      defence   \n",
+       "27             5          0        Control          got   \n",
+       "28             5          1           Pass        cross   \n",
+       "30             5          3      Challenge    challenge   \n",
+       "31             5          4          Field         edge   \n",
+       "33             5          6        Control     latching   \n",
+       "34             5          7           Shot       lofted   \n",
+       "35             5          8           Pass         ball   \n",
+       "36             5         10  Shot_Supports         shot   \n",
+       "37             5         11           Shot      whisker   \n",
+       "\n",
+       "                          changed_roles  \\\n",
+       "0                             [Rangers]   \n",
+       "3                             [Rangers]   \n",
+       "5                         [Man. United]   \n",
+       "6                         [Man. United]   \n",
+       "11                        [Man. United]   \n",
+       "12                            [Rangers]   \n",
+       "13                            [Rangers]   \n",
+       "14                        [Man. United]   \n",
+       "15                        [Man. United]   \n",
+       "16                        [Man. United]   \n",
+       "18  [Man. United, Rangers, Man. United]   \n",
+       "20                            [Rangers]   \n",
+       "25                            [Rangers]   \n",
+       "26                        [Man. United]   \n",
+       "27      [Rangers, Man. United, Rangers]   \n",
+       "28                            [Rangers]   \n",
+       "30                        [Man. United]   \n",
+       "31                        [Man. United]   \n",
+       "33                            [Rangers]   \n",
+       "34                        [Man. United]   \n",
+       "35                        [Man. United]   \n",
+       "36                        [Man. United]   \n",
+       "37                        [Man. United]   \n",
+       "\n",
+       "                                                roles  time_point   first_role  \n",
+       "0   [(COMPETITION_STAGE, [from, a, tight, Anglo-Sc...    0.000000      Rangers  \n",
+       "3   [(CONCEDING_TEAM, [they]), (CONCEDING_TEAM, [W...    0.187500      Rangers  \n",
+       "5   [(SUBSTITUTED_PLAYER, [Antonio, Valencia]), (T...    0.229167  Man. United  \n",
+       "6         [(SUBSTITUTED_PLAYER, [Antonio, Valencia])]    0.291667  Man. United  \n",
+       "11       [(MATCH_LOCATION, [at, Leeds, United, AFC])]    0.433333  Man. United  \n",
+       "12  [(COACH, [his]), (PLAYER, [of, three, central,...    0.466667      Rangers  \n",
+       "13  [(PLAYER, [Steven, Whittaker, and, Kirk, Broad...    0.500000      Rangers  \n",
+       "14  [(TEAM1, [United, 's]), (MATCH_LOCATION, [at, ...    0.523810  Man. United  \n",
+       "15  [(MOVING_BALL, [Fabio, 's, right-wing, cross])...    0.547619  Man. United  \n",
+       "16    [(TARGET, [right-wing]), (PASSER, [Fabio, 's])]    0.571429  Man. United  \n",
+       "18  [(MOVING_BALL, [Fabio, 's, right-wing, cross])...    0.619048  Man. United  \n",
+       "20  [(SHOT, [the, threat]), (INTERVENING_PLAYER, [...    0.666667      Rangers  \n",
+       "25                     [(RECIPIENT, [Kenny, Miller])]    0.796296      Rangers  \n",
+       "26                             [(TEAM, [United, 's])]    0.833333  Man. United  \n",
+       "27  [(RECIPIENT, [Kirk, Broadfoot]), (RECIPIENT, [...    0.833333      Rangers  \n",
+       "28                     [(PASSER, [Steven, Naismith])]    0.846154      Rangers  \n",
+       "30                    [(OPPONENT_PLAYER, [Smalling])]    0.871795  Man. United  \n",
+       "31                                 [(TEAM, [United])]    0.884615  Man. United  \n",
+       "33                   [(RECIPIENT, [Kirk, Broadfoot])]    0.910256      Rangers  \n",
+       "34  [(TARGET, [over, the, top]), (SHOOTER, [Ryan, ...    0.923077  Man. United  \n",
+       "35  [(TARGET, [over, the, top]), (PASSER, [Ryan, G...    0.935897  Man. United  \n",
+       "36                              [(SHOOTER, [Gibson])]    0.961538  Man. United  \n",
+       "37            [(TARGET, [over]), (SHOOTER, [Gibson])]    0.974359  Man. United  "
+      ]
+     },
+     "execution_count": 64,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_with_time = data_.assign(\n",
+    "    time_point= (data_\n",
+    "        .apply(lambda row: (row[\"sentence_idx\"] + row[\"frame_idx\"] / (sent_to_max_frame[row[\"sentence_idx\"]])) / (max_sent + 1), axis=1)\n",
+    "    )\n",
+    ")\n",
+    "\n",
+    "\n",
+    "data_with_first_roles = data_with_time.assign(\n",
+    "    first_role = data_with_time[\"changed_roles\"].apply(lambda roles: roles[0] if len(roles) > 0 else None)\n",
+    ")\n",
+    "\n",
+    "data_with_first_roles.dropna(axis=0, subset=[\"first_role\"])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 66,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Man. United    14\n",
+       "Rangers         9\n",
+       "Name: first_role, dtype: int64"
+      ]
+     },
+     "execution_count": 66,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_with_first_roles.dropna(axis=0, subset=[\"first_role\"])[\"first_role\"].value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 76,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'Intervene': 'Shot',\n",
+       " 'Challenge': 'One_On_One',\n",
+       " 'Award_Goal': 'Goal',\n",
+       " 'Deny': 'One_On_One',\n",
+       " 'Intercept': 'Pass',\n",
+       " 'Shot_Supports': 'Shot',\n",
+       " 'Save': 'Shot',\n",
+       " 'Pass': 'Pass',\n",
+       " 'Ball_Bounce': 'Motion',\n",
+       " 'Mark': 'Pass',\n",
+       " 'Goal_Kickoff': 'Shot',\n",
+       " 'Finish': 'Shot',\n",
+       " 'Beat': 'One_On_One',\n",
+       " 'Offside': 'Foul',\n",
+       " 'Bad_Pass': 'Pass',\n",
+       " 'Goal': 'Goal',\n",
+       " 'Set_Piece': 'Foul',\n",
+       " 'Shot': 'Shot',\n",
+       " 'Player_Move_With_Ball': 'Motion',\n",
+       " 'Sanction': 'Foul',\n",
+       " 'Move': 'Move',\n",
+       " 'Control': 'Pass',\n",
+       " 'Shoot_At': 'Shot',\n",
+       " 'Start_End_Match': 'Match',\n",
+       " 'Match': 'Match',\n",
+       " 'Deploy': 'Lineup',\n",
+       " 'Start': 'Lineup',\n",
+       " 'Player_Move': 'Motion',\n",
+       " 'Elimination': 'Match',\n",
+       " 'Result': 'Match',\n",
+       " 'Bring_Off': 'Substitution',\n",
+       " 'Away_Game': 'Match',\n",
+       " 'Possession': 'State_Of_Match',\n",
+       " 'Pass_Combination': 'Pass',\n",
+       " 'Lose_Ball': 'One_On_One',\n",
+       " 'Defense_Shot': 'Mix-Up',\n",
+       " 'Receive_Card': 'Foul',\n",
+       " 'Victory': 'Match',\n",
+       " 'Supply_Pass': 'Pass',\n",
+       " 'Bring_On': 'Substitution',\n",
+       " 'Chance': 'Chance',\n",
+       " 'Multiple_Goals': 'Goal',\n",
+       " 'One_On_One': 'One_On_One',\n",
+       " 'Substitute': 'Substitution',\n",
+       " 'Own_Goal': 'Goal',\n",
+       " 'Convert_Chance': 'Goal',\n",
+       " 'Ball_Land': 'Motion',\n",
+       " 'Progression': 'Match',\n",
+       " 'Home_Game': 'Match',\n",
+       " 'Referee_Decision': 'Foul',\n",
+       " 'Concede_Goal': 'Goal',\n",
+       " 'Mistake': 'Mix-Up',\n",
+       " 'Tactics': 'State_Of_Match',\n",
+       " 'Foul': 'Foul',\n",
+       " 'Being_Free': 'Pass',\n",
+       " 'Lead': 'State_Of_Match',\n",
+       " 'Create_Chance': 'Chance',\n",
+       " 'Win_Compensation': 'Foul',\n",
+       " 'Trail': 'State_Of_Match',\n",
+       " 'Ball_Move': 'Motion',\n",
+       " 'Defeat': 'Match',\n",
+       " 'Dissent': 'Foul',\n",
+       " 'Follow_Up': 'Shot',\n",
+       " 'Score': 'State_Of_Match',\n",
+       " 'Draw': 'Match',\n",
+       " 'Pass_Back': 'Pass',\n",
+       " 'Simulation': 'Foul',\n",
+       " 'Miss_Goal': 'Shot',\n",
+       " 'Miss_Chance': 'Chance',\n",
+       " 'Flick_On': 'Pass',\n",
+       " 'Ball_Escape': 'Motion',\n",
+       " 'Give_Card': 'Foul',\n",
+       " 'Take_On': 'One_On_One',\n",
+       " 'Connect': 'Pass',\n",
+       " 'Goalkeeper_Advance': 'Motion',\n",
+       " 'Hit': 'Shot',\n",
+       " 'Advantage': 'Foul',\n",
+       " 'Feign': 'Shot',\n",
+       " 'Overcome_Goalkeeper': 'Goal',\n",
+       " 'Prepare_Goal': 'Goal',\n",
+       " 'Ball': 'Actors',\n",
+       " 'Spectator_Activity': 'State_Of_Match',\n",
+       " 'Goal_Target': 'Field',\n",
+       " 'Player': 'Actors',\n",
+       " 'Referee': 'Actors',\n",
+       " 'Field': 'Field',\n",
+       " 'Team': 'Actors',\n",
+       " 'Bench': 'Field',\n",
+       " 'Coach': 'Actors',\n",
+       " 'Concede_Compensation': 'Foul',\n",
+       " 'Match_Temporal_Subdivision': 'Match',\n",
+       " 'Confusion': 'Mix-up',\n",
+       " 'Score_Goal': 'Goal',\n",
+       " 'Celebrate_Goal': 'Goal',\n",
+       " 'Suspension': 'Lineup',\n",
+       " 'Spectators': 'Actors',\n",
+       " 'Competition_Stage': 'Competition',\n",
+       " 'Body_Parts': 'Actors',\n",
+       " 'Trick': 'One_On_One',\n",
+       " 'Competition': 'Competition',\n",
+       " 'Stadium': 'Field',\n",
+       " 'Match_Quality': 'State_Of_Match',\n",
+       " 'Ball_And_Goal': 'Shot',\n",
+       " 'Equipment': 'Actors',\n",
+       " 'Injuries': 'Foul',\n",
+       " '': 'Pass',\n",
+       " 'Shot_Quality': 'Shot'}"
+      ]
+     },
+     "execution_count": 76,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "from lxml import etree as ET\n",
+    "kicktionary = ET.parse(\"kicktionary_lu_info.xml\")\n",
+    "frame_to_scenario = {\n",
+    "    lu.attrib[\"frame\"]: lu.attrib[\"scenario\"]\n",
+    "    for lu in kicktionary.xpath(\".//LEXICAL-UNIT\") if lu.attrib[\"frame\"]\n",
+    "}\n",
+    "frame_to_scenario"
+   ]
+  }
+ ],
+ "metadata": {
+  "interpreter": {
+   "hash": "1c3e697aa4e6ae3f4f7b554c053a671f50a0156c07fd90d57948174b29e29a61"
+  },
+  "kernelspec": {
+   "display_name": "Python 3.9.7 ('perspredict')",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.7"
+  },
+  "orig_nbformat": 4
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}