nsfwapp/backend/rating.go

// backend\rating.go

package main

import (
	"math"
	"strings"
)

type aiRatingMeta struct {
	Score                  float64 `json:"score"`
	Stars                  int     `json:"stars"`
	Segments               int     `json:"segments"`
	SegmentsPerMinute      float64 `json:"segmentsPerMinute"`
	FlaggedSeconds         float64 `json:"flaggedSeconds"`
	WeightedFlaggedSeconds float64 `json:"weightedFlaggedSeconds"`
	CoverageRatio          float64 `json:"coverageRatio"`
	WeightedCoverageRatio  float64 `json:"weightedCoverageRatio"`
	LongestSegmentSeconds  float64 `json:"longestSegmentSeconds"`
	AvgConfidence          float64 `json:"avgConfidence"`
}

func ratingClamp01(v float64) float64 {
	if v < 0 {
		return 0
	}
	if v > 1 {
		return 1
	}
	return v
}

func ratingRound(v float64, places int) float64 {
	p := math.Pow(10, float64(places))
	return math.Round(v*p) / p
}

func ratingSmoothStep(v float64) float64 {
	v = ratingClamp01(v)
	return v * v * (3 - 2*v)
}

func ratingSoftCap(value, knee float64) float64 {
	if value <= 0 || knee <= 0 {
		return 0
	}
	return value / (value + knee)
}

func ratingEffectiveDurationSeconds(seconds float64) float64 {
	if seconds <= 0 {
		return 0
	}

	// Lange Segmente zählen weiter, aber mit abnehmendem Zusatznutzen.
	// Dadurch kippen falsche oder zu breite Merges nicht sofort auf 5 Sterne.
	const kneeSeconds = 24.0
	return kneeSeconds * math.Log1p(seconds/kneeSeconds)
}

func nsfwSegmentSeverityWeight(label string) float64 {
	switch strings.ToLower(strings.TrimSpace(label)) {
	case "female_genitalia_exposed":
		return 1.00
	case "anus_exposed":
		return 0.95
	case "female_breast_exposed":
		return 0.85
	case "male_genitalia_exposed":
		return 0.80
	case "buttocks_exposed":
		return 0.65
	default:
		return 0.50
	}
}

func ratingConfidenceWeight(conf float64) float64 {
	conf = ratingClamp01(conf)

	// Unterhalb ~0.30 soll Confidence kaum boosten.
	// Oberhalb ~0.95 ist praktisch gesättigt.
	n := ratingSmoothStep((conf - 0.30) / 0.65)

	return 0.60 + 0.40*n
}

func starsFromNSFWScore(score float64) int {
	switch {
	case score < 15:
		return 1
	case score < 35:
		return 2
	case score < 58:
		return 3
	case score < 78:
		return 4
	default:
		return 5
	}
}

func computeNSFWRating(segments []aiSegmentMeta, durationSec float64) *aiRatingMeta {
	r := &aiRatingMeta{
		Score: 0,
		Stars: 1,
	}

	if durationSec <= 0 || len(segments) == 0 {
		return r
	}

	videoMinutes := math.Max(durationSec/60.0, 0.25)

	var totalFlagged float64
	var totalWeighted float64
	var totalEffectiveWeighted float64
	var longest float64
	var confSum float64
	var n int

	for _, s := range segments {
		segDur := s.DurationSeconds
		if segDur <= 0 {
			segDur = s.EndSeconds - s.StartSeconds
		}
		if segDur <= 0 {
			continue
		}

		sev := nsfwSegmentSeverityWeight(s.Label)
		conf := ratingClamp01(s.Score)
		weight := sev * ratingConfidenceWeight(conf)

		totalFlagged += segDur
		totalWeighted += segDur * weight
		totalEffectiveWeighted += ratingEffectiveDurationSeconds(segDur) * weight

		confSum += conf
		n++

		if segDur > longest {
			longest = segDur
		}
	}

	if n == 0 {
		return r
	}

	coverageRatio := ratingClamp01(totalFlagged / durationSec)
	weightedCoverageRatio := ratingClamp01(totalWeighted / durationSec)
	segmentsPerMinute := float64(n) / videoMinutes
	avgConfidence := confSum / float64(n)
	effectiveWeightedSecondsPerMinute := totalEffectiveWeighted / videoMinutes

	// Weichere Normalisierung statt harter Sättigung.
	// Das reduziert 1/5-Ausreißer und verteilt mehr Fälle auf 2–4 Sterne.
	densityNorm := ratingSoftCap(effectiveWeightedSecondsPerMinute, 8.0)
	coverageNorm := ratingSoftCap(weightedCoverageRatio, 0.22)
	frequencyNorm := ratingSoftCap(segmentsPerMinute, 1.25)
	longestNorm := ratingSoftCap(longest, 25.0)
	confNorm := ratingSmoothStep((avgConfidence - 0.35) / 0.60)

	raw :=
		0.38*densityNorm +
			0.30*coverageNorm +
			0.12*frequencyNorm +
			0.10*longestNorm +
			0.10*confNorm

	score := ratingRound(ratingClamp01(raw)*100, 1)

	r.Score = score
	r.Stars = starsFromNSFWScore(score)
	r.Segments = n
	r.SegmentsPerMinute = ratingRound(segmentsPerMinute, 2)
	r.FlaggedSeconds = ratingRound(totalFlagged, 2)
	r.WeightedFlaggedSeconds = ratingRound(totalWeighted, 2)
	r.CoverageRatio = ratingRound(coverageRatio, 4)
	r.WeightedCoverageRatio = ratingRound(weightedCoverageRatio, 4)
	r.LongestSegmentSeconds = ratingRound(longest, 2)
	r.AvgConfidence = ratingRound(avgConfidence, 3)

	return r
}