{ "cells": [ { "cell_type": "markdown", "id": "c76b8d41-0e76-47b4-97bb-b97ae0d96729", "metadata": {}, "source": [ "# 14 Sentiment Analysis" ] }, { "cell_type": "markdown", "id": "b0f76ae1-dfd9-4aa2-b6ca-ec104361692b", "metadata": {}, "source": [ "There are 2 common methods to perform sentiment analysis in NLP\n", "\n", "- pre-labeled training: a large dataset is given from wich to understand the sentiment of a new text.\n", "- VADER (Valence Aware Dictionary for sEntiment Reasoning): pre-trained model sensitive to both polarity (positive/negative) and intensity (strength) of an emotion." ] }, { "cell_type": "markdown", "id": "6c66c820-7d64-4fb3-ad2f-70796c5dee16", "metadata": {}, "source": [ "## VADER\n", "\n", "It relies on a dictionary wich maps lexical features to emotion intensities called sentiment score.\n", "\n", "The sentiment score of a text can be obtained by summing up the intensities of each word in the text." ] }, { "cell_type": "markdown", "id": "c4c580b3-dfe8-43a5-b6fa-a0307bf7ee6c", "metadata": {}, "source": [ "Sentiment Analysis on raw text is always challenging however, due to variety of possible factors:\n", "- Positive and Negative sentiment in the same text data\n", "- Sarcasm using positive words in a negative way" ] }, { "cell_type": "markdown", "id": "47f11a6f-4cad-4f37-a95a-decc447e59e6", "metadata": {}, "source": [ "## VADER with NLTK" ] }, { "cell_type": "code", "execution_count": 3, "id": "5ed7a5de-c65c-401b-88f2-f72296dfbb37", "metadata": {}, "outputs": [], "source": [ "import nltk" ] }, { "cell_type": "code", "execution_count": 4, "id": "f3035541-0386-4530-a791-f53ea7ab10b8", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "[nltk_data] Downloading package vader_lexicon to\n", "[nltk_data] /home/jovyan/nltk_data...\n" ] }, { "data": { "text/plain": [ "True" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "nltk.download('vader_lexicon')" ] }, { "cell_type": "code", "execution_count": 5, "id": "d8dd3af6-9de1-4bca-868d-b490ce75469a", "metadata": {}, "outputs": [], "source": [ "from nltk.sentiment.vader import SentimentIntensityAnalyzer" ] }, { "cell_type": "code", "execution_count": 6, "id": "93a4029e-5d95-4dbf-aa00-ee465f8babe9", "metadata": {}, "outputs": [], "source": [ "sid = SentimentIntensityAnalyzer()" ] }, { "cell_type": "code", "execution_count": 7, "id": "60c90301-9f80-427e-aa50-3ca3c8b0e04c", "metadata": {}, "outputs": [], "source": [ "a = \"This is a good movie\"" ] }, { "cell_type": "code", "execution_count": 8, "id": "7663dfbe-acd6-4911-8435-35c5c6ae48a7", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'neg': 0.0, 'neu': 0.508, 'pos': 0.492, 'compound': 0.4404}" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "sid.polarity_scores(a)" ] }, { "cell_type": "code", "execution_count": 9, "id": "c0b4842f-b7d3-4785-bde5-2ba5b6098abf", "metadata": {}, "outputs": [], "source": [ "a = \"This was the best, most awesome movie EVER MADE!!!\"" ] }, { "cell_type": "code", "execution_count": 10, "id": "2791d1d7-dcea-47a0-915e-78e869e7edbf", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'neg': 0.0, 'neu': 0.425, 'pos': 0.575, 'compound': 0.8877}" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "sid.polarity_scores(a)" ] }, { "cell_type": "code", "execution_count": 11, "id": "a3ec8a13-971c-4cfe-8165-08e904efdf2f", "metadata": {}, "outputs": [], "source": [ "a = \"This was the WORST movie that has ever disgraced the screen.\"" ] }, { "cell_type": "code", "execution_count": 12, "id": "de9f9fb0-88a1-4c13-afe3-4b33de29dbd9", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'neg': 0.465, 'neu': 0.535, 'pos': 0.0, 'compound': -0.8331}" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "sid.polarity_scores(a)" ] }, { "cell_type": "markdown", "id": "1d72f6b5-2de2-4e02-ba72-3846225b70db", "metadata": {}, "source": [ "## Amazon Reviews Project" ] }, { "cell_type": "code", "execution_count": 13, "id": "18531223-664c-4d10-87af-bf8347560c1c", "metadata": {}, "outputs": [], "source": [ "import pandas as pd" ] }, { "cell_type": "code", "execution_count": 18, "id": "8369c3d3-3707-4716-9b69-d8facab54441", "metadata": {}, "outputs": [], "source": [ "df = pd.read_csv('amazonreviews.tsv',sep='\\t')" ] }, { "cell_type": "code", "execution_count": 19, "id": "1570a95a-607b-41e3-80c6-35c6025e5b5f", "metadata": {}, "outputs": [ { "data": { "text/html": [ "

	label	review
0	pos	Stuning even for the non-gamer: This sound tra...
1	pos	The best soundtrack ever to anything.: I'm rea...
2	pos	Amazing!: This soundtrack is my favorite music...
3	pos	Excellent Soundtrack: I truly like this soundt...
4	pos	Remember, Pull Your Jaw Off The Floor After He...

\n", "

" ], "text/plain": [ " label review\n", "0 pos Stuning even for the non-gamer: This sound tra...\n", "1 pos The best soundtrack ever to anything.: I'm rea...\n", "2 pos Amazing!: This soundtrack is my favorite music...\n", "3 pos Excellent Soundtrack: I truly like this soundt...\n", "4 pos Remember, Pull Your Jaw Off The Floor After He..." ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.head()" ] }, { "cell_type": "code", "execution_count": 20, "id": "4995960c-6c54-427b-af7b-07b4a3de96af", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "label\n", "neg 5097\n", "pos 4903\n", "Name: count, dtype: int64" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df['label'].value_counts()" ] }, { "cell_type": "code", "execution_count": 21, "id": "c4ed8b9c-495a-460f-9c16-4ab25f9a6479", "metadata": {}, "outputs": [], "source": [ "df.dropna(inplace=True)" ] }, { "cell_type": "code", "execution_count": 23, "id": "efb6b7e9-46f9-4452-afcd-8bef59c788c4", "metadata": {}, "outputs": [], "source": [ "blanks = []\n", "for i,label,review in df.itertuples():\n", " if type(review) == str:\n", " if review.isspace():\n", " blanks.append(i)\n", "df.drop(blanks, inplace=True)" ] }, { "cell_type": "code", "execution_count": 25, "id": "f003bb9a-31d4-4911-a60f-2c258e91c2ca", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'Stuning even for the non-gamer: This sound track was beautiful! It paints the senery in your mind so well I would recomend it even to people who hate vid. game music! I have played the game Chrono Cross but out of all of the games I have ever played it has the best music! It backs away from crude keyboarding and takes a fresher step with grate guitars and soulful orchestras. It would impress anyone who cares to listen! ^_^'" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.iloc[0]['review']" ] }, { "cell_type": "code", "execution_count": 24, "id": "d35ef5c7-f627-44e4-b474-5ee6815583d8", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'neg': 0.088, 'neu': 0.669, 'pos': 0.243, 'compound': 0.9454}" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "sid.polarity_scores(df.iloc[0]['review'])" ] }, { "cell_type": "code", "execution_count": 27, "id": "df1c75ad-47f7-48ae-a46e-ee8e9efaac1d", "metadata": {}, "outputs": [], "source": [ "df['scores'] = df['review'].apply(lambda review: sid.polarity_scores(review))" ] }, { "cell_type": "code", "execution_count": 29, "id": "81b49b92-7faa-4ea5-8170-57eee2dcb85d", "metadata": {}, "outputs": [ { "data": { "text/html": [ "

\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "

	label	review	scores
0	pos	Stuning even for the non-gamer: This sound tra...	{'neg': 0.088, 'neu': 0.669, 'pos': 0.243, 'co...
1	pos	The best soundtrack ever to anything.: I'm rea...	{'neg': 0.018, 'neu': 0.837, 'pos': 0.145, 'co...
2	pos	Amazing!: This soundtrack is my favorite music...	{'neg': 0.04, 'neu': 0.692, 'pos': 0.268, 'com...
3	pos	Excellent Soundtrack: I truly like this soundt...	{'neg': 0.09, 'neu': 0.615, 'pos': 0.295, 'com...
4	pos	Remember, Pull Your Jaw Off The Floor After He...	{'neg': 0.0, 'neu': 0.746, 'pos': 0.254, 'comp...

\n", "

	label	review	scores	compound
0	pos	Stuning even for the non-gamer: This sound tra...	{'neg': 0.088, 'neu': 0.669, 'pos': 0.243, 'co...	0.9454
1	pos	The best soundtrack ever to anything.: I'm rea...	{'neg': 0.018, 'neu': 0.837, 'pos': 0.145, 'co...	0.8957
2	pos	Amazing!: This soundtrack is my favorite music...	{'neg': 0.04, 'neu': 0.692, 'pos': 0.268, 'com...	0.9858
3	pos	Excellent Soundtrack: I truly like this soundt...	{'neg': 0.09, 'neu': 0.615, 'pos': 0.295, 'com...	0.9814
4	pos	Remember, Pull Your Jaw Off The Floor After He...	{'neg': 0.0, 'neu': 0.746, 'pos': 0.254, 'comp...	0.9781

\n", "

	label	review	scores	compound	comp_score
0	pos	Stuning even for the non-gamer: This sound tra...	{'neg': 0.088, 'neu': 0.669, 'pos': 0.243, 'co...	0.9454	pos
1	pos	The best soundtrack ever to anything.: I'm rea...	{'neg': 0.018, 'neu': 0.837, 'pos': 0.145, 'co...	0.8957	pos
2	pos	Amazing!: This soundtrack is my favorite music...	{'neg': 0.04, 'neu': 0.692, 'pos': 0.268, 'com...	0.9858	pos
3	pos	Excellent Soundtrack: I truly like this soundt...	{'neg': 0.09, 'neu': 0.615, 'pos': 0.295, 'com...	0.9814	pos
4	pos	Remember, Pull Your Jaw Off The Floor After He...	{'neg': 0.0, 'neu': 0.746, 'pos': 0.254, 'comp...	0.9781	pos

\n", "

" ], "text/plain": [ " label review \\\n", "0 pos Stuning even for the non-gamer: This sound tra... \n", "1 pos The best soundtrack ever to anything.: I'm rea... \n", "2 pos Amazing!: This soundtrack is my favorite music... \n", "3 pos Excellent Soundtrack: I truly like this soundt... \n", "4 pos Remember, Pull Your Jaw Off The Floor After He... \n", "\n", " scores compound comp_score \n", "0 {'neg': 0.088, 'neu': 0.669, 'pos': 0.243, 'co... 0.9454 pos \n", "1 {'neg': 0.018, 'neu': 0.837, 'pos': 0.145, 'co... 0.8957 pos \n", "2 {'neg': 0.04, 'neu': 0.692, 'pos': 0.268, 'com... 0.9858 pos \n", "3 {'neg': 0.09, 'neu': 0.615, 'pos': 0.295, 'com... 0.9814 pos \n", "4 {'neg': 0.0, 'neu': 0.746, 'pos': 0.254, 'comp... 0.9781 pos " ] }, "execution_count": 33, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.head()" ] }, { "cell_type": "code", "execution_count": 34, "id": "db916c41-c25e-445b-87d4-bab20c4238c8", "metadata": {}, "outputs": [], "source": [ "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix" ] }, { "cell_type": "code", "execution_count": 35, "id": "c7938b97-2928-4077-b86c-babfb9d113eb", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.7097" ] }, "execution_count": 35, "metadata": {}, "output_type": "execute_result" } ], "source": [ "accuracy_score(df['label'], df['comp_score'])" ] }, { "cell_type": "code", "execution_count": 36, "id": "50c276cb-6f85-4828-a0c8-c2418dcabb09", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " precision recall f1-score support\n", "\n", " neg 0.86 0.52 0.64 5097\n", " pos 0.64 0.91 0.75 4903\n", "\n", " accuracy 0.71 10000\n", " macro avg 0.75 0.71 0.70 10000\n", "weighted avg 0.75 0.71 0.70 10000\n", "\n" ] } ], "source": [ "print(classification_report(df['label'],df['comp_score']))" ] }, { "cell_type": "code", "execution_count": 37, "id": "a854ef4d-352f-4d3f-8955-fbd225c58684", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[[2629 2468]\n", " [ 435 4468]]\n" ] } ], "source": [ "print(confusion_matrix(df['label'],df['comp_score']))" ] }, { "cell_type": "markdown", "id": "f106912e-fb2e-4c3b-9ff3-9a8efd251ce1", "metadata": {}, "source": [ "## Another Project" ] }, { "cell_type": "code", "execution_count": 40, "id": "86a9bd44-8715-4427-a826-5541d7454600", "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd" ] }, { "cell_type": "code", "execution_count": 43, "id": "4d7231bb-906a-4f36-a42a-602dadbe7962", "metadata": {}, "outputs": [], "source": [ "df = pd.read_csv('moviereviews.tsv', sep='\\t')" ] }, { "cell_type": "code", "execution_count": 44, "id": "c7f13d36-4f64-4787-8f9f-34e8eda5032a", "metadata": {}, "outputs": [ { "data": { "text/html": [ "

	label	review
0	neg	how do films like mouse hunt get into theatres...
1	neg	some talented actresses are blessed with a dem...
2	pos	this has been an extraordinary year for austra...
3	pos	according to hollywood movies made in last few...
4	neg	my first press screening of 1998 and already i...

\n", "

" ], "text/plain": [ " label review\n", "0 neg how do films like mouse hunt get into theatres...\n", "1 neg some talented actresses are blessed with a dem...\n", "2 pos this has been an extraordinary year for austra...\n", "3 pos according to hollywood movies made in last few...\n", "4 neg my first press screening of 1998 and already i..." ] }, "execution_count": 44, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.head()" ] }, { "cell_type": "code", "execution_count": 45, "id": "f048d7dd-489e-4c93-9b40-baa8dea4153a", "metadata": {}, "outputs": [], "source": [ "df.dropna(inplace=True)" ] }, { "cell_type": "code", "execution_count": 46, "id": "c3287a0b-60f9-4c73-b072-361a12aba5ec", "metadata": {}, "outputs": [], "source": [ "blanks = []\n", "for i,label,review in df.itertuples():\n", " if type(review) == str:\n", " if review.isspace():\n", " blanks.append(i)\n", "df.drop(blanks, inplace=True)" ] }, { "cell_type": "code", "execution_count": 47, "id": "8a186855-ff4c-4ab8-84d5-5f47e1b5850a", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "label\n", "neg 969\n", "pos 969\n", "Name: count, dtype: int64" ] }, "execution_count": 47, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df['label'].value_counts()" ] }, { "cell_type": "code", "execution_count": 49, "id": "168b9784-ae37-4bc3-9231-443d25b92289", "metadata": {}, "outputs": [], "source": [ "from nltk.sentiment.vader import SentimentIntensityAnalyzer" ] }, { "cell_type": "code", "execution_count": 50, "id": "d76b8366-2c0c-446d-b444-c58d5628f544", "metadata": {}, "outputs": [], "source": [ "sid = SentimentIntensityAnalyzer()" ] }, { "cell_type": "code", "execution_count": 51, "id": "028be615-526d-4e33-93bd-e831633dda42", "metadata": {}, "outputs": [], "source": [ "df['scores'] = df['review'].apply(lambda review:sid.polarity_scores(review))" ] }, { "cell_type": "code", "execution_count": 52, "id": "1e62f669-7c8b-46eb-b8d8-e3c7524bc5be", "metadata": {}, "outputs": [], "source": [ "df['compound'] = df['scores'].apply(lambda d:d['compound'])" ] }, { "cell_type": "code", "execution_count": 53, "id": "a4003993-85d6-4b98-85da-9dd5c8c58c32", "metadata": {}, "outputs": [ { "data": { "text/html": [ "

	label	review	scores	compound
0	neg	how do films like mouse hunt get into theatres...	{'neg': 0.121, 'neu': 0.778, 'pos': 0.101, 'co...	-0.9125
1	neg	some talented actresses are blessed with a dem...	{'neg': 0.12, 'neu': 0.775, 'pos': 0.105, 'com...	-0.8618
2	pos	this has been an extraordinary year for austra...	{'neg': 0.068, 'neu': 0.781, 'pos': 0.15, 'com...	0.9951
3	pos	according to hollywood movies made in last few...	{'neg': 0.071, 'neu': 0.782, 'pos': 0.147, 'co...	0.9972
4	neg	my first press screening of 1998 and already i...	{'neg': 0.091, 'neu': 0.817, 'pos': 0.093, 'co...	-0.2484

\n", "

	label	review	scores	compound	comp_score
0	neg	how do films like mouse hunt get into theatres...	{'neg': 0.121, 'neu': 0.778, 'pos': 0.101, 'co...	-0.9125	neg
1	neg	some talented actresses are blessed with a dem...	{'neg': 0.12, 'neu': 0.775, 'pos': 0.105, 'com...	-0.8618	neg
2	pos	this has been an extraordinary year for austra...	{'neg': 0.068, 'neu': 0.781, 'pos': 0.15, 'com...	0.9951	pos
3	pos	according to hollywood movies made in last few...	{'neg': 0.071, 'neu': 0.782, 'pos': 0.147, 'co...	0.9972	pos
4	neg	my first press screening of 1998 and already i...	{'neg': 0.091, 'neu': 0.817, 'pos': 0.093, 'co...	-0.2484	neg

\n", "

" ], "text/plain": [ " label review \\\n", "0 neg how do films like mouse hunt get into theatres... \n", "1 neg some talented actresses are blessed with a dem... \n", "2 pos this has been an extraordinary year for austra... \n", "3 pos according to hollywood movies made in last few... \n", "4 neg my first press screening of 1998 and already i... \n", "\n", " scores compound comp_score \n", "0 {'neg': 0.121, 'neu': 0.778, 'pos': 0.101, 'co... -0.9125 neg \n", "1 {'neg': 0.12, 'neu': 0.775, 'pos': 0.105, 'com... -0.8618 neg \n", "2 {'neg': 0.068, 'neu': 0.781, 'pos': 0.15, 'com... 0.9951 pos \n", "3 {'neg': 0.071, 'neu': 0.782, 'pos': 0.147, 'co... 0.9972 pos \n", "4 {'neg': 0.091, 'neu': 0.817, 'pos': 0.093, 'co... -0.2484 neg " ] }, "execution_count": 57, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.head()" ] }, { "cell_type": "code", "execution_count": 58, "id": "1deedd16-c059-4511-a7f8-97dd47b1e8cd", "metadata": {}, "outputs": [], "source": [ "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix" ] }, { "cell_type": "code", "execution_count": 59, "id": "b39190af-14b3-4241-9f78-d3951689cd4f", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.6357069143446853" ] }, "execution_count": 59, "metadata": {}, "output_type": "execute_result" } ], "source": [ "accuracy_score(df['label'], df['comp_score'])" ] }, { "cell_type": "code", "execution_count": 60, "id": "0c5bc7c7-12cc-454c-ac66-bcd74795c0a7", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "' precision recall f1-score support\\n\\n neg 0.72 0.44 0.55 969\\n pos 0.60 0.83 0.70 969\\n\\n accuracy 0.64 1938\\n macro avg 0.66 0.64 0.62 1938\\nweighted avg 0.66 0.64 0.62 1938\\n'" ] }, "execution_count": 60, "metadata": {}, "output_type": "execute_result" } ], "source": [ "classification_report(df['label'], df['comp_score'])" ] }, { "cell_type": "code", "execution_count": 61, "id": "547e2c93-baa9-4538-8515-dde6b9dc0ea6", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([[427, 542],\n", " [164, 805]])" ] }, "execution_count": 61, "metadata": {}, "output_type": "execute_result" } ], "source": [ "confusion_matrix(df['label'], df['comp_score'])" ] }, { "cell_type": "code", "execution_count": null, "id": "653450e3-e07a-46db-817c-33ee38730556", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.11.4" } }, "nbformat": 4, "nbformat_minor": 5 }