Added README

README.md

@@ -0,0 +1,7 @@
+# Natural Language Processing with Amazon Reviews
+
+## Setup & Installation
+
+1. Install python3 and the necessary packages
+2. To run `proProcessBagOfWords.ipynb`, you need to [download the spreadsheet files](https://www.kaggle.com/datasets/yacharki/amazon-reviews-for-sentianalysis-finegrained-csv) that exceed GitHub's limit.
+3. There are some `test` and `train` spreadsheet files already generated if you want to skip step 2.

amazonNLP.ipynb

@@ -308,15 +308,13 @@
     "    # Train with allWords\n",
     "    clf.fit(X_train, y_train, **arg)\n",
     "\n",
-    "    grid = HalvingGridSearchCV(clf, param_grid) # If we want Cross Validation: cv=pdfsplt\n",
+    "    grid = HalvingGridSearchCV(clf, param_grid)\n",
     "\n",
     "    grid.fit(X_paramTuning, y_paramTuning, **arg)\n",
     "\n",
     "    print(\"Best grid params:\", grid.best_params_)\n",
     "\n",
-    "    return grid.predict(testAllWords)\n",
-    "\n",
-    "# MultinomialNB()"
+    "    return grid.predict(testAllWords)\n"
    ]
   },
   {
@@ -351,7 +349,7 @@
     "            currData = data[index]\n",
     "            x, y, title = prepare_statistic(currData)\n",
     "            axis = axs[i, j]\n",
-    "            axis.set_ylim(bottom=0.0, top=1.0)    # Uncomment this to see yAxis from [0, 1]\n",
+    "            axis.set_ylim(bottom=0.0, top=1.0)\n",
     "            axis.bar(x, y)\n",
     "            axis.set_title(title)\n",
     "            axis.grid(True)\n",
@@ -521,8 +519,8 @@
     "begin = time.time()\n",
     "prediction = predict(MultinomialNB(), sample_weight=1.0)\n",
     "end = time.time()\n",
-    "# prediction = gridPredict(MultinomialNB(), {\"alpha\": [0.0001, 0.5, 1.0]}, sample_weight=1.0)\n",
     "# prediction = fullPredict(MultinomialNB(), {\"alpha\": [0.0001, 0.5, 1.0]}, sample_weight=1.0)\n",
+    "\n",
     "print(\"MultinomialNB Results:\")\n",
     "accuracy = analyze(prediction, test, 'Multinomial Naive Bayes')\n",
     "nb_metrics = (accuracy, end - begin)"