Reorganized similarity code

usc-isi-i2 · Jan 15, 2022 · e98df72 · e98df72
1 parent 9a7986d
commit e98df72
Show file tree

Hide file tree

Showing 30 changed files with 135,830 additions and 14,134 deletions.
diff --git a/.ipynb_checkpoints/Basis Datasets Generation-checkpoint.ipynb b/.ipynb_checkpoints/Basis Datasets Generation-checkpoint.ipynb
diff --git a/.ipynb_checkpoints/ConceptNet Evaluation Dataset Exploration-checkpoint.ipynb b/.ipynb_checkpoints/ConceptNet Evaluation Dataset Exploration-checkpoint.ipynb
@@ -1,9 +1,17 @@
 {
  "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "derived-behalf",
+   "metadata": {},
+   "source": [
+    "In this notebook, we explore the properties of conceptnet as an evaluation benchmark"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 1,
-   "id": "amended-investment",
+   "id": "hidden-bathroom",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -13,7 +21,7 @@
   {
    "cell_type": "code",
    "execution_count": 2,
-   "id": "mature-demonstration",
+   "id": "changing-methodology",
    "metadata": {},
    "outputs": [
     {
@@ -48,7 +56,7 @@
   {
    "cell_type": "code",
    "execution_count": 5,
-   "id": "dense-holly",
+   "id": "incoming-things",
    "metadata": {},
    "outputs": [
     {
@@ -66,7 +74,7 @@
   {
    "cell_type": "code",
    "execution_count": 6,
-   "id": "linear-chaos",
+   "id": "solar-brain",
    "metadata": {},
    "outputs": [
     {
@@ -104,7 +112,7 @@
   {
    "cell_type": "code",
    "execution_count": 7,
-   "id": "broken-nightlife",
+   "id": "adjusted-artist",
    "metadata": {},
    "outputs": [
     {
@@ -123,7 +131,7 @@
   {
    "cell_type": "code",
    "execution_count": 8,
-   "id": "conceptual-timer",
+   "id": "phantom-communication",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -140,7 +148,7 @@
   {
    "cell_type": "code",
    "execution_count": 9,
-   "id": "practical-witch",
+   "id": "headed-choice",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -151,7 +159,7 @@
   {
    "cell_type": "code",
    "execution_count": 10,
-   "id": "tracked-academy",
+   "id": "cognitive-potential",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -161,7 +169,7 @@
   {
    "cell_type": "code",
    "execution_count": 11,
-   "id": "fatty-overhead",
+   "id": "charming-momentum",
    "metadata": {},
    "outputs": [
     {
@@ -179,7 +187,7 @@
   {
    "cell_type": "code",
    "execution_count": 12,
-   "id": "presidential-circus",
+   "id": "european-porter",
    "metadata": {},
    "outputs": [
     {
@@ -209,7 +217,7 @@
   {
    "cell_type": "code",
    "execution_count": 13,
-   "id": "mobile-monday",
+   "id": "several-convention",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -219,7 +227,7 @@
   {
    "cell_type": "code",
    "execution_count": 14,
-   "id": "pursuant-bullet",
+   "id": "separate-fortune",
    "metadata": {},
    "outputs": [
     {
@@ -237,7 +245,7 @@
   {
    "cell_type": "code",
    "execution_count": 15,
-   "id": "passive-snake",
+   "id": "straight-health",
    "metadata": {},
    "outputs": [
     {
@@ -294,7 +302,7 @@
   {
    "cell_type": "code",
    "execution_count": 20,
-   "id": "psychological-equilibrium",
+   "id": "apparent-pointer",
    "metadata": {},
    "outputs": [
     {
@@ -314,10 +322,33 @@
     "conceptDF2['category'] = conceptDF2['relation;label'].apply(lambda p: 'I' if p == 'synonym' or p == 'similar to' else 'M')"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "id": "sapphire-control",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "M    10220\n",
+       "I     2175\n",
+       "Name: relation;label, dtype: int64"
+      ]
+     },
+     "execution_count": 22,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "conceptDF2['relation;label'].apply(lambda p: 'I' if p == 'synonym' or p == 'similar to' else 'M').value_counts()"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 18,
-   "id": "cosmetic-monday",
+   "id": "daily-strain",
    "metadata": {},
    "outputs": [
     {
@@ -347,7 +378,7 @@
   {
    "cell_type": "code",
    "execution_count": 21,
-   "id": "grave-tooth",
+   "id": "meaningful-aquatic",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -384,7 +415,7 @@
    "toc_cell": false,
    "toc_position": {},
    "toc_section_display": true,
-   "toc_window_display": false
+   "toc_window_display": true
   },
   "varInspector": {
    "cols": {