marek/MtoRagSystem
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

optimize intents

Browse Source
This commit is contained in:
team2
2026-02-27 07:12:36 +01:00
parent 26ee95cadd
commit 2713da1afb
2 changed files with 155 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

144
src/Intent/IntentLite.php Normal file
View File

@@ -0,0 +1,144 @@
<?php
declare(strict_types=1);
namespace App\Intent;
/**
* IntentLite
*
* Deterministische, LLM-agnostische Intent-Erkennung.
* Fokus: LIST-Intent für Retrieval-Steuerung.
*
* WICHTIG:
* - Immer mit dem ORIGINAL-Prompt aufrufen.
* - Nicht mit dem QueryCleaner-Ergebnis.
*/
final class IntentLite
{
private const LIST_THRESHOLD = 4;
public function detectList(string $originalPrompt): array
{
$p = $this->normalize($originalPrompt);
$score = 0;
$signals = [];
// --------------------------------------------------------
// 1. Starke explizite Listen-Trigger (hohes Gewicht)
// --------------------------------------------------------
$strongPatterns = [
'/\bliste(n)?\b/u',
'/\bauflisten\b/u',
'/\baufz(a|ä)hl(en)?\b/u',
'/\bnenn(e)?\b/u',
'/\bzeig(e)?\b/u',
'/\bwelche\s+sind\b/u',
'/\bwelche\s+gibt\s+es\b/u',
'/\bwas\s+sind\b/u',
'/\bwie\s+viele\b/u',
'/\branking\b/u',
'/\btop\s*\d+\b/u',
];
foreach ($strongPatterns as $pattern) {
if (preg_match($pattern, $p) === 1) {
$score += 3;
$signals[] = "strong:$pattern";
}
}
// --------------------------------------------------------
// 2. Mengen- / Mehrzahl-Indikatoren
// --------------------------------------------------------
$quantityWords = [
'alle',
'sämtliche',
'saemtliche',
'mehrere',
'verschiedene',
'einige',
'viele',
'optionen',
'möglichkeiten',
'moeglichkeiten',
'varianten',
'arten',
'modelle',
'funktionen',
'punkte',
'schritte',
'kategorien',
'übersicht',
'uebersicht',
];
foreach ($quantityWords as $word) {
if (preg_match('/\b' . preg_quote($word, '/') . '\b/u', $p) === 1) {
$score += 2;
$signals[] = "quantity:$word";
}
}
// --------------------------------------------------------
// 3. Explizite Zahlen (z. B. "5 Vorteile")
// --------------------------------------------------------
if (preg_match('/\b\d+\b/u', $p) === 1) {
$score += 2;
$signals[] = 'number';
}
// --------------------------------------------------------
// 4. Enumeration-Hinweise (1., -, *, etc.)
// --------------------------------------------------------
if (
preg_match('/(^|\s)(\d+\)|\d+\.|-\s|\*\s)/u', $originalPrompt) === 1
) {
$score += 1;
$signals[] = 'enumeration_hint';
}
// --------------------------------------------------------
// Entscheidung
// --------------------------------------------------------
$isList = $score >= self::LIST_THRESHOLD;
return [
'is_list' => $isList,
'score' => $score,
'signals' => $signals,
];
}
public function isListQuery(string $originalPrompt): bool
{
return $this->detectList($originalPrompt)['is_list'];
}
// ------------------------------------------------------------
// Interne Normalisierung (ohne Stopword-Entfernung!)
// ------------------------------------------------------------
private function normalize(string $s): string
{
$s = mb_strtolower($s);
// Umlaute zusätzlich absichern (falls QueryCleaner das tut)
$replacements = [
'ä' => 'ae',
'ö' => 'oe',
'ü' => 'ue',
'ß' => 'ss',
];
// Nur als Zusatzform speichern (nicht ersetzen!)
foreach ($replacements as $umlaut => $alt) {
if (str_contains($s, $umlaut)) {
$s .= ' ' . str_replace($umlaut, $alt, $s);
break;
}
}
return $s;
}
}

21
src/Knowledge/Retrieval/NdjsonHybridRetriever.php
View File

@@ -5,6 +5,7 @@ declare(strict_types=1);
namespace App\Knowledge\Retrieval;
use App\Entity\ModelGenerationConfig;
use App\Intent\IntentLite;
use App\Knowledge\ChunkManager;
use App\Knowledge\QueryCleaner;
use App\Repository\ModelGenerationConfigRepository;
@@ -30,7 +31,10 @@ final class NdjsonHybridRetriever implements RetrieverInterface
private readonly TagRoutingService $tagRouting,
private readonly ModelGenerationConfigRepository $configRepository,
private readonly QueryCleaner $queryCleaner,
) {}
private readonly IntentLite $intentLite
)
{
}
public function retrieve(string $prompt): array
{
@@ -54,7 +58,7 @@ final class NdjsonHybridRetriever implements RetrieverInterface
$vectorTopKBase = max(1, min($config->getRetrievalVectorTopK(), self::HARD_MAX_VECTORK));
// Wichtig: List-Detection bleibt auf Originalprompt (sonst entfernst du "zeige/liste" etc.)
$isListQuery = $this->isListQuery($prompt);
$isListQuery = $this->intentLite->isListQuery($prompt);
// -------------------------------------------------
// CLEAN QUERY (nur für Retrieval: Tags + Vector)
@@ -81,7 +85,7 @@ final class NdjsonHybridRetriever implements RetrieverInterface
// List mode: höhere Abdeckung, um mehr Dokumente zu ranken
if ($isListQuery) {
$topK = max($vectorTopKBase * 3, 80);
$topK = (int)round($vectorTopKBase * 2.5);
}
$topK = max(1, min($topK, self::HARD_MAX_VECTORK));
@@ -194,17 +198,6 @@ final class NdjsonHybridRetriever implements RetrieverInterface
// LIST QUERY DETECTION
// =========================================================
private function isListQuery(string $prompt): bool
{
$prompt = mb_strtolower($prompt);
return str_contains($prompt, 'liste')
|| str_contains($prompt, 'zeige')
|| str_contains($prompt, 'nenn')
|| str_contains($prompt, 'welche')
|| preg_match('/\b\d+\b/', $prompt) === 1;
}
// =========================================================
// DOCUMENT RANKING (Adjusted Scores incl. Tag Bonus)
// =========================================================