From b6ed452fc6fc3c893e02384220292f772af674ca Mon Sep 17 00:00:00 2001 From: Jonas Weinz Date: Tue, 5 Jun 2018 15:03:33 +0200 Subject: [PATCH] updated naive approach --- Project/naive_approach/naive_approach.ipynb | 237 ++++++++++---------- 1 file changed, 119 insertions(+), 118 deletions(-) diff --git a/Project/naive_approach/naive_approach.ipynb b/Project/naive_approach/naive_approach.ipynb index 0e4205e..c508e17 100644 --- a/Project/naive_approach/naive_approach.ipynb +++ b/Project/naive_approach/naive_approach.ipynb @@ -247,11 +247,12 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 21, "metadata": {}, "outputs": [], "source": [ - "def evaluate_sentence(sentence, description_key = 'description'):\n", + "def evaluate_sentence(sentence, description_key = 'description', lang = 'eng'):\n", + " \n", " tokenized_sentence = word_tokenize(sentence)\n", " n = len(tokenized_sentence)\n", " l = table.shape[0]\n", @@ -264,12 +265,12 @@ " mat = np.zeros(shape=(m,n))\n", " for i in range(len(emoji_tokens)):\n", " for j in range(len(tokenized_sentence)):\n", - " syn1 = wordnet.synsets(emoji_tokens[i])\n", + " syn1 = wordnet.synsets(emoji_tokens[i],lang=lang)\n", " if len(syn1) == 0:\n", " continue\n", " w1 = syn1[0]\n", " #print(j, tokenized_sentence)\n", - " syn2 = wordnet.synsets(tokenized_sentence[j])\n", + " syn2 = wordnet.synsets(tokenized_sentence[j], lang=lang)\n", " if len(syn2) == 0:\n", " continue\n", " w2 = syn2[0]\n", @@ -294,7 +295,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 22, "metadata": {}, "outputs": [], "source": [ @@ -316,13 +317,13 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 23, "metadata": {}, "outputs": [], "source": [ - "def predict(sentence, description_key='description', n=10, t=0.9):\n", + "def predict(sentence, description_key='description', lang = 'eng', n=10, t=0.9):\n", "\n", - " result = evaluate_sentence(sentence, description_key)\n", + " result = evaluate_sentence(sentence, description_key, lang)\n", " \n", " summed = np.argsort([-np.sum(x) for x in result])[0:n]\n", " max_val = np.argsort([-np.max(x) for x in result])[0:n]\n", @@ -343,7 +344,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 29, "metadata": {}, "outputs": [ { @@ -380,113 +381,113 @@ " \n", " \n", " 0\n", - " ๐Ÿคช\n", - " GRINNING FACE WITH ONE LARGE AND ONE SMALL EYE\n", + " ๐Ÿคญ\n", + " SMILING FACE WITH SMILING EYES AND HAND COVERI...\n", " ๐ŸคŸ\n", " I LOVE YOU HAND SIGN\n", - " ๐Ÿ’ข\n", - " ANGER SYMBOL\n", - " ๐Ÿ’Œ\n", - " LOVE LETTER\n", + " ๐Ÿ’\n", + " BOUQUET\n", + " ๐Ÿš†\n", + " TRAIN\n", " \n", " \n", " 1\n", - " ๐Ÿšฎ\n", - " PUT LITTER IN ITS PLACE SYMBOL\n", - " ๐Ÿ’Œ\n", - " LOVE LETTER\n", - " ๐Ÿ“ฆ\n", - " PACKAGE\n", - " ๐Ÿฉ\n", - " LOVE HOTEL\n", + " ๐Ÿ”›\n", + " ON WITH EXCLAMATION MARK WITH LEFT RIGHT ARROW...\n", + " ๐Ÿ‡ฎ\n", + " REGIONAL INDICATOR SYMBOL LETTER I\n", + " ๐Ÿณ\n", + " COOKING\n", + " ๐Ÿšธ\n", + " CHILDREN CROSSING\n", " \n", " \n", " 2\n", - " ๐Ÿ––\n", - " RAISED HAND WITH PART BETWEEN MIDDLE AND RING ...\n", - " ๐Ÿ‡ฎ\n", - " REGIONAL INDICATOR SYMBOL LETTER I\n", - " ๐Ÿ’Œ\n", - " LOVE LETTER\n", - " ๐ŸคŸ\n", - " I LOVE YOU HAND SIGN\n", + " ๐Ÿšฎ\n", + " PUT LITTER IN ITS PLACE SYMBOL\n", + " ๐Ÿš†\n", + " TRAIN\n", + " ๐Ÿš†\n", + " TRAIN\n", + " ๐Ÿš„\n", + " HIGH-SPEED TRAIN\n", " \n", " \n", " 3\n", - " ๐Ÿฅฐ\n", - " SMILING FACE WITH SMILING EYES AND THREE HEARTS\n", - " ๐Ÿฉ\n", - " LOVE HOTEL\n", - " ๐Ÿ†‘\n", - " SQUARED CL\n", + " ๐Ÿคช\n", + " GRINNING FACE WITH ONE LARGE AND ONE SMALL EYE\n", + " ๐Ÿš…\n", + " HIGH-SPEED TRAIN WITH BULLET NOSE\n", + " ๐ŸŽฅ\n", + " MOVIE CAMERA\n", " ๐Ÿ‡ฎ\n", " REGIONAL INDICATOR SYMBOL LETTER I\n", " \n", " \n", " 4\n", - " ๐Ÿคญ\n", - " SMILING FACE WITH SMILING EYES AND HAND COVERI...\n", - " ๐Ÿ˜ฑ\n", - " FACE SCREAMING IN FEAR\n", - " ๐ŸŒ‰\n", - " BRIDGE AT NIGHT\n", - " ๐Ÿ“จ\n", - " INCOMING ENVELOPE\n", + " ๐Ÿฅฐ\n", + " SMILING FACE WITH SMILING EYES AND THREE HEARTS\n", + " ๐Ÿš„\n", + " HIGH-SPEED TRAIN\n", + " ๐ŸŽญ\n", + " PERFORMING ARTS\n", + " ๐ŸคŸ\n", + " I LOVE YOU HAND SIGN\n", " \n", " \n", " 5\n", - " ๐Ÿ”‚\n", - " CLOCKWISE RIGHTWARDS AND LEFTWARDS OPEN CIRCLE...\n", - " ๐Ÿ‡ญ\n", - " REGIONAL INDICATOR SYMBOL LETTER H\n", - " ๐Ÿ’\n", - " BOUQUET\n", - " ๐Ÿ“ฉ\n", - " ENVELOPE WITH DOWNWARDS ARROW ABOVE\n", + " ๐Ÿ™Œ\n", + " PERSON RAISING BOTH HANDS IN CELEBRATION\n", + " ๐Ÿšธ\n", + " CHILDREN CROSSING\n", + " ๐ŸŽน\n", + " MUSICAL KEYBOARD\n", + " ๐Ÿš…\n", + " HIGH-SPEED TRAIN WITH BULLET NOSE\n", " \n", " \n", " 6\n", - " ๐Ÿ”›\n", - " ON WITH EXCLAMATION MARK WITH LEFT RIGHT ARROW...\n", - " ๐Ÿ‡ณ\n", - " REGIONAL INDICATOR SYMBOL LETTER N\n", - " ๐Ÿ“œ\n", - " SCROLL\n", - " ๐Ÿ“ช\n", - " CLOSED MAILBOX WITH LOWERED FLAG\n", + " ๐Ÿ––\n", + " RAISED HAND WITH PART BETWEEN MIDDLE AND RING ...\n", + " ๐Ÿ›ธ\n", + " FLYING SAUCER\n", + " ๐Ÿงพ\n", + " RECEIPT\n", + " ๐Ÿ”น\n", + " SMALL BLUE DIAMOND\n", " \n", " \n", " 7\n", - " ๐Ÿ” \n", - " INPUT SYMBOL FOR LATIN CAPITAL LETTERS\n", - " ๐Ÿ‡ด\n", - " REGIONAL INDICATOR SYMBOL LETTER O\n", - " ๐ŸŽ\n", - " WRAPPED PRESENT\n", - " ๐Ÿ“ซ\n", - " CLOSED MAILBOX WITH RAISED FLAG\n", + " ๐Ÿ”‚\n", + " CLOCKWISE RIGHTWARDS AND LEFTWARDS OPEN CIRCLE...\n", + " ๐Ÿฅพ\n", + " HIKING BOOT\n", + " ๐Ÿ’\n", + " KISS\n", + " ๐Ÿ“ \n", + " FAX MACHINE\n", " \n", " \n", " 8\n", - " ๐ŸคŸ\n", - " I LOVE YOU HAND SIGN\n", - " ๐Ÿ“ฒ\n", - " MOBILE PHONE WITH RIGHTWARDS ARROW AT LEFT\n", - " ๐Ÿ’จ\n", - " DASH SYMBOL\n", - " ๐Ÿ“ฐ\n", - " NEWSPAPER\n", + " ๐Ÿ˜„\n", + " SMILING FACE WITH OPEN MOUTH AND SMILING EYES\n", + " ๐Ÿฅ\n", + " FLYING DISC\n", + " ๐Ÿ”ฅ\n", + " FIRE\n", + " ๐Ÿ“ก\n", + " SATELLITE ANTENNA\n", " \n", " \n", " 9\n", - " ๐Ÿ‡ฎ\n", - " REGIONAL INDICATOR SYMBOL LETTER I\n", - " ๐ŸŒ†\n", - " CITYSCAPE AT DUSK\n", - " ๐Ÿ’ค\n", - " SLEEPING SYMBOL\n", - " ๐Ÿ“ฌ\n", - " OPEN MAILBOX WITH RAISED FLAG\n", + " ๐Ÿ‘‰\n", + " WHITE RIGHT POINTING BACKHAND INDEX\n", + " ๐Ÿš\n", + " BUS STOP\n", + " ๐ŸŽผ\n", + " MUSICAL SCORE\n", + " ๐Ÿ“ข\n", + " PUBLIC ADDRESS LOUDSPEAKER\n", " \n", " \n", "\n", @@ -494,40 +495,40 @@ ], "text/plain": [ " summed summed_description max_val \\\n", - "0 ๐Ÿคช GRINNING FACE WITH ONE LARGE AND ONE SMALL EYE ๐ŸคŸ \n", - "1 ๐Ÿšฎ PUT LITTER IN ITS PLACE SYMBOL ๐Ÿ’Œ \n", - "2 ๐Ÿ–– RAISED HAND WITH PART BETWEEN MIDDLE AND RING ... ๐Ÿ‡ฎ \n", - "3 ๐Ÿฅฐ SMILING FACE WITH SMILING EYES AND THREE HEARTS ๐Ÿฉ \n", - "4 ๐Ÿคญ SMILING FACE WITH SMILING EYES AND HAND COVERI... ๐Ÿ˜ฑ \n", - "5 ๐Ÿ”‚ CLOCKWISE RIGHTWARDS AND LEFTWARDS OPEN CIRCLE... ๐Ÿ‡ญ \n", - "6 ๐Ÿ”› ON WITH EXCLAMATION MARK WITH LEFT RIGHT ARROW... ๐Ÿ‡ณ \n", - "7 ๐Ÿ”  INPUT SYMBOL FOR LATIN CAPITAL LETTERS ๐Ÿ‡ด \n", - "8 ๐ŸคŸ I LOVE YOU HAND SIGN ๐Ÿ“ฒ \n", - "9 ๐Ÿ‡ฎ REGIONAL INDICATOR SYMBOL LETTER I ๐ŸŒ† \n", + "0 ๐Ÿคญ SMILING FACE WITH SMILING EYES AND HAND COVERI... ๐ŸคŸ \n", + "1 ๐Ÿ”› ON WITH EXCLAMATION MARK WITH LEFT RIGHT ARROW... ๐Ÿ‡ฎ \n", + "2 ๐Ÿšฎ PUT LITTER IN ITS PLACE SYMBOL ๐Ÿš† \n", + "3 ๐Ÿคช GRINNING FACE WITH ONE LARGE AND ONE SMALL EYE ๐Ÿš… \n", + "4 ๐Ÿฅฐ SMILING FACE WITH SMILING EYES AND THREE HEARTS ๐Ÿš„ \n", + "5 ๐Ÿ™Œ PERSON RAISING BOTH HANDS IN CELEBRATION ๐Ÿšธ \n", + "6 ๐Ÿ–– RAISED HAND WITH PART BETWEEN MIDDLE AND RING ... ๐Ÿ›ธ \n", + "7 ๐Ÿ”‚ CLOCKWISE RIGHTWARDS AND LEFTWARDS OPEN CIRCLE... ๐Ÿฅพ \n", + "8 ๐Ÿ˜„ SMILING FACE WITH OPEN MOUTH AND SMILING EYES ๐Ÿฅ \n", + "9 ๐Ÿ‘‰ WHITE RIGHT POINTING BACKHAND INDEX ๐Ÿš \n", "\n", - " max_val_description avg avg_description threshold \\\n", - "0 I LOVE YOU HAND SIGN ๐Ÿ’ข ANGER SYMBOL ๐Ÿ’Œ \n", - "1 LOVE LETTER ๐Ÿ“ฆ PACKAGE ๐Ÿฉ \n", - "2 REGIONAL INDICATOR SYMBOL LETTER I ๐Ÿ’Œ LOVE LETTER ๐ŸคŸ \n", - "3 LOVE HOTEL ๐Ÿ†‘ SQUARED CL ๐Ÿ‡ฎ \n", - "4 FACE SCREAMING IN FEAR ๐ŸŒ‰ BRIDGE AT NIGHT ๐Ÿ“จ \n", - "5 REGIONAL INDICATOR SYMBOL LETTER H ๐Ÿ’ BOUQUET ๐Ÿ“ฉ \n", - "6 REGIONAL INDICATOR SYMBOL LETTER N ๐Ÿ“œ SCROLL ๐Ÿ“ช \n", - "7 REGIONAL INDICATOR SYMBOL LETTER O ๐ŸŽ WRAPPED PRESENT ๐Ÿ“ซ \n", - "8 MOBILE PHONE WITH RIGHTWARDS ARROW AT LEFT ๐Ÿ’จ DASH SYMBOL ๐Ÿ“ฐ \n", - "9 CITYSCAPE AT DUSK ๐Ÿ’ค SLEEPING SYMBOL ๐Ÿ“ฌ \n", + " max_val_description avg avg_description threshold \\\n", + "0 I LOVE YOU HAND SIGN ๐Ÿ’ BOUQUET ๐Ÿš† \n", + "1 REGIONAL INDICATOR SYMBOL LETTER I ๐Ÿณ COOKING ๐Ÿšธ \n", + "2 TRAIN ๐Ÿš† TRAIN ๐Ÿš„ \n", + "3 HIGH-SPEED TRAIN WITH BULLET NOSE ๐ŸŽฅ MOVIE CAMERA ๐Ÿ‡ฎ \n", + "4 HIGH-SPEED TRAIN ๐ŸŽญ PERFORMING ARTS ๐ŸคŸ \n", + "5 CHILDREN CROSSING ๐ŸŽน MUSICAL KEYBOARD ๐Ÿš… \n", + "6 FLYING SAUCER ๐Ÿงพ RECEIPT ๐Ÿ”น \n", + "7 HIKING BOOT ๐Ÿ’ KISS ๐Ÿ“  \n", + "8 FLYING DISC ๐Ÿ”ฅ FIRE ๐Ÿ“ก \n", + "9 BUS STOP ๐ŸŽผ MUSICAL SCORE ๐Ÿ“ข \n", "\n", - " threshold_description \n", - "0 LOVE LETTER \n", - "1 LOVE HOTEL \n", - "2 I LOVE YOU HAND SIGN \n", - "3 REGIONAL INDICATOR SYMBOL LETTER I \n", - "4 INCOMING ENVELOPE \n", - "5 ENVELOPE WITH DOWNWARDS ARROW ABOVE \n", - "6 CLOSED MAILBOX WITH LOWERED FLAG \n", - "7 CLOSED MAILBOX WITH RAISED FLAG \n", - "8 NEWSPAPER \n", - "9 OPEN MAILBOX WITH RAISED FLAG " + " threshold_description \n", + "0 TRAIN \n", + "1 CHILDREN CROSSING \n", + "2 HIGH-SPEED TRAIN \n", + "3 REGIONAL INDICATOR SYMBOL LETTER I \n", + "4 I LOVE YOU HAND SIGN \n", + "5 HIGH-SPEED TRAIN WITH BULLET NOSE \n", + "6 SMALL BLUE DIAMOND \n", + "7 FAX MACHINE \n", + "8 SATELLITE ANTENNA \n", + "9 PUBLIC ADDRESS LOUDSPEAKER " ] }, "metadata": {}, @@ -535,7 +536,7 @@ } ], "source": [ - "predict(\"I love sunny days!\", description_key='description' )" + "predict(\"I like to travel by train\", description_key='description' , lang='eng')" ] }, {