simple chat classification

diff --git a/app/Models/ChatLog.php b/app/Models/ChatLog.php
index 8c890e707112acc3ade9ea8450b821def2c92915..f918269073a0fa733e7ea9765a0a57d514b2bca2 100644 (file)
--- a/app/Models/ChatLog.php
+++ b/app/Models/ChatLog.php
@@ -4,6 +4,8 @@ namespace App\Models;
 
 use Illuminate\Database\Eloquent\Factories\HasFactory;
 use Illuminate\Database\Eloquent\Model;
+use Illuminate\Support\Arr;
+use Illuminate\Support\Str;
 use LanguageDetector\LanguageDetector;
 
 class ChatLog extends Model {
@@ -44,6 +46,7 @@ class ChatLog extends Model {
                        if ($this->scanForSpam()) {
                                $this->banned = true;
                        }
+                       $this->classification = static::classify($this->text_content);
                        return;
                }
 
@@ -64,6 +67,36 @@ class ChatLog extends Model {
                ]);
        }
 
+       public static function classify($text) {
+               if (empty($text)) {
+                       return 'unclassified';
+               }
+               if (is_numeric(trim($text))) {
+                       return 'number';
+               }
+               $rawText = strtolower(preg_replace('/[^\w]/', '', $text));
+               $tokenizedText = preg_split('/\s+/', strtolower(trim($text)));
+               if (Str::startsWith($rawText, 'gg') || Str::endsWith($rawText, 'gg')) {
+                       return 'gg';
+               }
+               if (Str::contains($rawText, ['glgl', 'glhf', 'hfgl'])) {
+                       return 'gl';
+               }
+               if (Str::contains($rawText, ['haha', 'hehe', 'hihi', 'kekw', 'lol', 'lul', 'xd'])) {
+                       return 'lol';
+               }
+               if (Str::startsWith($rawText, ['ahoi', 'hallo', 'hello', 'hi', 'huhu']) || Str::endsWith($rawText, ['hi', 'wave'])) {
+                       return 'hi';
+               }
+               if (Str::contains($rawText, ['pog', 'wow'])) {
+                       return 'pog';
+               }
+               if (Str::contains($rawText, ['hype'])) {
+                       return 'hype';
+               }
+               return 'unclassified';
+       }
+
        protected function evaluateUser() {
        }
 

diff --git a/database/migrations/2024_02_18_150400_chat_classification.php b/database/migrations/2024_02_18_150400_chat_classification.php
new file mode 100644 (file)
index 0000000..ad67e8d
--- /dev/null
+++ b/database/migrations/2024_02_18_150400_chat_classification.php
@@ -0,0 +1,33 @@
+<?php
+
+use Illuminate\Database\Migrations\Migration;
+use Illuminate\Database\Schema\Blueprint;
+use Illuminate\Support\Facades\Schema;
+
+return new class extends Migration
+{
+       /**
+        * Run the migrations.
+        *
+        * @return void
+        */
+       public function up()
+       {
+               Schema::table('chat_logs', function (Blueprint $table) {
+                       $table->string('classification')->default('unclassified');
+                       $table->index(['classification']);
+               });
+       }
+
+       /**
+        * Reverse the migrations.
+        *
+        * @return void
+        */
+       public function down()
+       {
+               Schema::table('chat_logs', function (Blueprint $table) {
+                       $table->dropColumn('classification');
+               });
+       }
+};