{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Subjectivity Analysis" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "\n", "This tutorial is available as an IPython notebook at [Malaya/example/subjectivity](https://github.com/huseinzol05/Malaya/tree/master/example/subjectivity).\n", " \n", "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "\n", "This module trained on both standard and local (included social media) language structures, so it is save to use for both.\n", " \n", "
" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import logging\n", "\n", "logging.basicConfig(level=logging.INFO)" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "INFO:numexpr.utils:NumExpr defaulting to 8 threads.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "CPU times: user 5.2 s, sys: 752 ms, total: 5.95 s\n", "Wall time: 5.58 s\n" ] } ], "source": [ "%%time\n", "import malaya" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### labels supported\n", "\n", "Default labels for subjectivity module." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['negative', 'positive']" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "malaya.subjectivity.label" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Explanation\n", "\n", "Positive subjectivity: based on or influenced by personal feelings, tastes, or opinions. Can be a positive or negative sentiment.\n", "\n", "Negative subjectivity: based on a report or a fact. Can be a positive or negative sentiment." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "negative_text = 'Kerajaan negeri Kelantan mempersoalkan motif kenyataan Menteri Kewangan Lim Guan Eng yang hanya menyebut Kelantan penerima terbesar bantuan kewangan dari Kerajaan Persekutuan. Sedangkan menurut Timbalan Menteri Besarnya, Datuk Mohd Amar Nik Abdullah, negeri lain yang lebih maju dari Kelantan turut mendapat pembiayaan dan pinjaman.'\n", "positive_text = 'kerajaan sebenarnya sangat bencikan rakyatnya, minyak naik dan segalanya'\n", "\n", "string1 = 'Sis, students from overseas were brought back because they are not in their countries which is if something happens to them, its not the other countries’ responsibility. Student dalam malaysia ni dah dlm tggjawab kerajaan. Mana part yg tak faham?'\n", "string2 = 'Harap kerajaan tak bukak serentak. Slowly release week by week. Focus on economy related industries dulu'\n", "string3 = 'Idk if aku salah baca ke apa. Bayaran rm350 utk golongan umur 21 ke bawah shj ? Anyone? If 21 ke atas ok lah. If umur 21 ke bawah? Are you serious? Siapa yg lebih byk komitmen? Aku hrp aku salah baca. Aku tk jumpa artikel tu'\n", "string4 = 'Jabatan Penjara Malaysia diperuntukkan RM20 juta laksana program pembangunan Insan kepada banduan. Majikan yang menggaji bekas banduan, bekas penagih dadah diberi potongan cukai tambahan sehingga 2025.'" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Load multinomial model\n", "\n", "```python\n", "def multinomial(**kwargs):\n", " \"\"\"\n", " Load multinomial emotion model.\n", "\n", " Returns\n", " -------\n", " result : malaya.model.ml.Bayes class\n", " \"\"\"\n", "```" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "model = malaya.subjectivity.multinomial()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Predict batch of strings\n", "\n", "```python\n", "def predict(self, strings: List[str], add_neutral: bool = True):\n", " \"\"\"\n", " classify list of strings.\n", "\n", " Parameters\n", " ----------\n", " strings: List[str]\n", " add_neutral: bool, optional (default=True)\n", " if True, it will add neutral probability.\n", "\n", " Returns\n", " -------\n", " result: List[str]\n", " \"\"\"\n", "```" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['neutral', 'negative']" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.predict([positive_text,negative_text])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Disable `neutral` probability," ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['positive', 'negative']" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.predict([positive_text,negative_text], add_neutral = False)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Predict batch of strings with probability\n", "\n", "```python\n", "def predict_proba(self, strings: List[str], add_neutral: bool = True):\n", " \"\"\"\n", " classify list of strings and return probability.\n", "\n", " Parameters\n", " ----------\n", " strings: List[str]\n", " add_neutral: bool, optional (default=True)\n", " if True, it will add neutral probability.\n", "\n", " Returns\n", " -------\n", " result: List[dict[str, float]]\n", " \"\"\"\n", "```" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[{'negative': 0.420659316666446, 'positive': 0.5793406833335559},\n", " {'negative': 0.7906212884104161, 'positive': 0.2093787115895868}]" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.predict_proba([positive_text,negative_text], add_neutral = False)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### List available Transformer models" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "INFO:malaya.subjectivity:trained on 80% dataset, tested on another 20% test set, dataset at https://github.com/huseinzol05/Malay-Dataset/tree/master/corpus/subjectivity\n" ] }, { "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", " \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", "
Size (MB)Quantized Size (MB)macro precisionmacro recallmacro f1-score
bert425.6111.000.920040.917480.91663
tiny-bert57.415.400.910230.902280.90301
albert48.612.800.905440.902990.90300
tiny-albert22.45.980.894570.894690.89461
xlnet446.6118.000.919160.917530.91761
alxlnet46.813.300.908620.908350.90817
fastformer458.0116.000.807850.819730.80758
tiny-fastformer77.319.700.871470.871470.87105
\n", "
" ], "text/plain": [ " Size (MB) Quantized Size (MB) macro precision \\\n", "bert 425.6 111.00 0.92004 \n", "tiny-bert 57.4 15.40 0.91023 \n", "albert 48.6 12.80 0.90544 \n", "tiny-albert 22.4 5.98 0.89457 \n", "xlnet 446.6 118.00 0.91916 \n", "alxlnet 46.8 13.30 0.90862 \n", "fastformer 458.0 116.00 0.80785 \n", "tiny-fastformer 77.3 19.70 0.87147 \n", "\n", " macro recall macro f1-score \n", "bert 0.91748 0.91663 \n", "tiny-bert 0.90228 0.90301 \n", "albert 0.90299 0.90300 \n", "tiny-albert 0.89469 0.89461 \n", "xlnet 0.91753 0.91761 \n", "alxlnet 0.90835 0.90817 \n", "fastformer 0.81973 0.80758 \n", "tiny-fastformer 0.87147 0.87105 " ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "malaya.subjectivity.available_transformer()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Load Transformer model\n", "\n", "All model interface will follow sklearn interface started v3.4,\n", "\n", "```python\n", "def transformer(model: str = 'bert', quantized: bool = False, **kwargs):\n", " \"\"\"\n", " Load Transformer subjectivity model.\n", "\n", " Parameters\n", " ----------\n", " model : str, optional (default='bert')\n", " Model architecture supported. Allowed values:\n", "\n", " * ``'bert'`` - Google BERT BASE parameters.\n", " * ``'tiny-bert'`` - Google BERT TINY parameters.\n", " * ``'albert'`` - Google ALBERT BASE parameters.\n", " * ``'tiny-albert'`` - Google ALBERT TINY parameters.\n", " * ``'xlnet'`` - Google XLNET BASE parameters.\n", " * ``'alxlnet'`` - Malaya ALXLNET BASE parameters.\n", " * ``'fastformer'`` - FastFormer BASE parameters.\n", " * ``'tiny-fastformer'`` - FastFormer TINY parameters.\n", "\n", " quantized : bool, optional (default=False)\n", " if True, will load 8-bit quantized model.\n", " Quantized model not necessary faster, totally depends on the machine.\n", "\n", " Returns\n", " -------\n", " result: model\n", " List of model classes:\n", "\n", " * if `bert` in model, will return `malaya.model.bert.BinaryBERT`.\n", " * if `xlnet` in model, will return `malaya.model.xlnet.BinaryXLNET`.\n", " * if `fastformer` in model, will return `malaya.model.fastformer.BinaryFastFormer`.\n", " \"\"\"\n", "```" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "101%|██████████| 49.0/48.6 [00:35<00:00, 1.39MB/s]\n", "184%|██████████| 1.00/0.54 [00:02<-1:59:59, 2.83s/MB]\n", "135%|██████████| 1.00/0.74 [00:03<00:00, 3.73s/MB]\n" ] } ], "source": [ "model = malaya.subjectivity.transformer(model = 'albert')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Load Quantized model\n", "\n", "To load 8-bit quantized model, simply pass `quantized = True`, default is `False`.\n", "\n", "We can expect slightly accuracy drop from quantized model, and not necessary faster than normal 32-bit float model, totally depends on machine." ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "WARNING:root:Load quantized model will cause accuracy drop.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "INFO:tensorflow:loading sentence piece model\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "INFO:tensorflow:loading sentence piece model\n" ] } ], "source": [ "quantized_model = malaya.subjectivity.transformer(model = 'albert', quantized = True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Predict batch of strings\n", "\n", "```python\n", "def predict(self, strings: List[str], add_neutral: bool = True):\n", " \"\"\"\n", " classify list of strings.\n", "\n", " Parameters\n", " ----------\n", " strings: List[str]\n", " add_neutral: bool, optional (default=True)\n", " if True, it will add neutral probability.\n", "\n", " Returns\n", " -------\n", " result: List[str]\n", " \"\"\"\n", "```" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['negative', 'negative']" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.predict([negative_text, positive_text])" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['negative', 'negative']" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "quantized_model.predict([negative_text, positive_text])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Predict batch of strings with probability\n", "\n", "```python\n", "def predict_proba(self, strings: List[str], add_neutral: bool = True):\n", " \"\"\"\n", " classify list of strings and return probability.\n", "\n", " Parameters\n", " ----------\n", " strings: List[str]\n", " add_neutral: bool, optional (default=True)\n", " if True, it will add neutral probability.\n", "\n", " Returns\n", " -------\n", " result: List[dict[str, float]]\n", " \"\"\"\n", "```" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[{'negative': 0.9956738, 'positive': 4.326162e-05, 'neutral': 0.0042829514},\n", " {'negative': 0.9615872, 'positive': 0.00038412912, 'neutral': 0.038028657}]" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.predict_proba([negative_text, positive_text])" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[{'negative': 0.9954784, 'positive': 4.521673e-05, 'neutral': 0.0044763684},\n", " {'negative': 0.9612684, 'positive': 0.00038731584, 'neutral': 0.038344264}]" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "quantized_model.predict_proba([negative_text, positive_text])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Open subjectivity visualization dashboard\n", "\n", "Default when you call `predict_words` it will open a browser with visualization dashboard, you can disable by `visualization=False`.\n", "\n", "```python\n", "def predict_words(\n", " self,\n", " string: str,\n", " method: str = 'last',\n", " bins_size: float = 0.05,\n", " visualization: bool = True,\n", "):\n", " \"\"\"\n", " classify words.\n", "\n", " Parameters\n", " ----------\n", " string : str\n", " method : str, optional (default='last')\n", " Attention layer supported. Allowed values:\n", "\n", " * ``'last'`` - attention from last layer.\n", " * ``'first'`` - attention from first layer.\n", " * ``'mean'`` - average attentions from all layers.\n", " bins_size: float, optional (default=0.05)\n", " default bins size for word distribution histogram.\n", " visualization: bool, optional (default=True)\n", " If True, it will open the visualization dashboard.\n", "\n", " Returns\n", " -------\n", " dictionary: results\n", " \"\"\"\n", "```\n" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "model.predict_words(negative_text)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Vectorize\n", "\n", "Let say you want to visualize sentence / word level in lower dimension, you can use `model.vectorize`,\n", "\n", "```python\n", "def vectorize(self, strings: List[str], method: str = 'first'):\n", " \"\"\"\n", " vectorize list of strings.\n", "\n", " Parameters\n", " ----------\n", " strings: List[str]\n", " method : str, optional (default='first')\n", " Vectorization layer supported. Allowed values:\n", "\n", " * ``'last'`` - vector from last sequence.\n", " * ``'first'`` - vector from first sequence.\n", " * ``'mean'`` - average vectors from all sequences.\n", " * ``'word'`` - average vectors based on tokens.\n", "\n", " Returns\n", " -------\n", " result: np.array\n", " \"\"\"\n", "```" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Sentence level" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "texts = [negative_text, positive_text, string1, string2]\n", "r = quantized_model.vectorize(texts, method = 'first')" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(4, 2)" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.manifold import TSNE\n", "import matplotlib.pyplot as plt\n", "\n", "tsne = TSNE().fit_transform(r)\n", "tsne.shape" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plt.figure(figsize = (7, 7))\n", "plt.scatter(tsne[:, 0], tsne[:, 1])\n", "labels = texts\n", "for label, x, y in zip(\n", " labels, tsne[:, 0], tsne[:, 1]\n", "):\n", " label = (\n", " '%s, %.3f' % (label[0], label[1])\n", " if isinstance(label, list)\n", " else label\n", " )\n", " plt.annotate(\n", " label,\n", " xy = (x, y),\n", " xytext = (0, 0),\n", " textcoords = 'offset points',\n", " )" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Word level" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "r = quantized_model.vectorize(texts, method = 'word')" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "x, y = [], []\n", "for row in r:\n", " x.extend([i[0] for i in row])\n", " y.extend([i[1] for i in row])" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(109, 2)" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tsne = TSNE().fit_transform(y)\n", "tsne.shape" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plt.figure(figsize = (7, 7))\n", "plt.scatter(tsne[:, 0], tsne[:, 1])\n", "labels = x\n", "for label, x, y in zip(\n", " labels, tsne[:, 0], tsne[:, 1]\n", "):\n", " label = (\n", " '%s, %.3f' % (label[0], label[1])\n", " if isinstance(label, list)\n", " else label\n", " )\n", " plt.annotate(\n", " label,\n", " xy = (x, y),\n", " xytext = (0, 0),\n", " textcoords = 'offset points',\n", " )" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Pretty good, the model able to know cluster top side as positive subjectivity, bottom side as negative subjectivity." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Stacking models\n", "\n", "More information, you can read at [https://malaya.readthedocs.io/en/latest/Stack.html](https://malaya.readthedocs.io/en/latest/Stack.html)" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "multinomial = malaya.subjectivity.multinomial()\n", "alxlnet = malaya.subjectivity.transformer(model = 'alxlnet')" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[{'negative': 0.19735892950073536,\n", " 'positive': 0.003119166818228667,\n", " 'neutral': 0.1160071232668102}]" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "malaya.stack.predict_stack([multinomial, model, alxlnet], [positive_text])" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[{'negative': 0.7424157666636825, 'positive': 0.04498033797670938}]" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "malaya.stack.predict_stack([multinomial, model, alxlnet], [positive_text], add_neutral = False)" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "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": "3.7.7" }, "varInspector": { "cols": { "lenName": 16, "lenType": 16, "lenVar": 40 }, "kernels_config": { "python": { "delete_cmd_postfix": "", "delete_cmd_prefix": "del ", "library": "var_list.py", "varRefreshCmd": "print(var_dic_list())" }, "r": { "delete_cmd_postfix": ") ", "delete_cmd_prefix": "rm(", "library": "var_list.r", "varRefreshCmd": "cat(var_dic_list()) " } }, "types_to_exclude": [ "module", "function", "builtin_function_or_method", "instance", "_Feature" ], "window_display": false } }, "nbformat": 4, "nbformat_minor": 2 }