Rebuild the coding keywords with 100 groups

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

runs/02-cluster/keywords-to-merge.json

@@ -1,55 +1,105 @@
 {
  "keywords": [
-  "abandonment-and-solitude",
+  "addiction-and-obsession",
   "alcohol-and-substance-abuse",
+  "ambition-and-self-determination",
   "attraction-and-admiration",
-  "betrayal-and-mistrust",
-  "boastful-self-confidence",
-  "breakup-and-reconciliation",
+  "authenticity-and-honesty",
+  "avoiding-commitment",
+  "betrayal-and-cheating",
+  "carefree-escapism",
   "celebration-and-partying",
-  "contentment-and-joy",
+  "cherishing-the-present-moment",
+  "clinging-to-love",
+  "community-and-togetherness",
+  "confidence-and-self-assurance",
+  "criminal-secrecy",
   "dancing-and-movement",
-  "defiance-and-confrontation",
-  "denial-and-pretense",
-  "designer-fashion-flexing",
-  "devotion-and-sacrifice",
-  "embracing-true-identity",
+  "deception-and-distrust",
+  "defiance-and-rebellion",
+  "denial-and-self-deception",
+  "desire-and-lust",
+  "devotion",
+  "disappointment-and-failure",
+  "dominance-and-power",
+  "dream-and-awakening",
+  "ex-moving-on",
+  "faith-and-spirituality",
+  "false-friendship",
   "fame-and-status",
+  "family-and-distance",
   "family-and-fatherhood",
-  "fantasy-and-imagination",
-  "fear-of-vulnerability",
+  "fear-of-losing-love",
   "female-empowerment",
   "flirtation-and-seduction",
   "freedom-and-escape",
+  "gratitude-and-appreciation",
+  "guilt-and-apology",
+  "haters-and-critics",
   "heartbreak-and-grief",
   "hidden-emotional-struggle",
-  "hope-and-disillusionment",
+  "hope-and-optimism",
   "human-connection",
-  "hustle-and-self-made-success",
+  "hustle-and-money",
+  "idealized-beauty",
+  "impermanence",
   "independence-and-self-reliance",
+  "individual-identity",
+  "inner-mental-turmoil",
   "intimacy-and-connection",
-  "jealousy-and-resentment-from-others",
-  "longing-and-desire",
+  "jealousy-and-bitterness",
+  "legacy-and-family",
+  "lingering-memories",
+  "loneliness-and-isolation",
+  "longing-and-loss",
   "longing-for-clarity",
+  "longing-for-reconciliation",
   "love-as-fleeting-and-transient",
-  "loyalty-and-commitment",
+  "loyalty-and-relationships",
+  "luxury-and-fashion",
+  "mortality-and-death",
+  "music-as-entertainment",
   "mutual-individuality",
-  "nostalgia-and-memory",
-  "obsession-and-madness",
+  "nighttime-revelry",
+  "non-judgment-and-acceptance",
+  "nostalgia-and-longing",
+  "overcoming-hardship",
   "partner-inadequacy",
   "partying-and-nightlife",
+  "past-trauma-and-healing",
   "personal-growth-and-change",
+  "possessiveness-and-exclusivity",
+  "predatory-threat",
+  "reassurance-and-comfort",
+  "reassurance-and-protection",
   "reggaeton-culture",
-  "regret-and-guilt",
+  "regret-and-reflection",
+  "rejection-of-women",
+  "relationship-breakup",
+  "relationship-complications",
   "relentless-ambition",
-  "resilience-through-hardship",
+  "reluctance-to-let-go",
+  "reputation-and-gossip",
+  "resilience-and-strength",
+  "rising-to-success",
   "rivalry-and-superiority",
   "romantic-pursuit",
-  "secrecy-and-paranoia",
+  "self-confidence-and-braggadocio",
   "self-worth-and-insecurity",
+  "sensual-desire",
+  "showing-off-and-impressing",
   "small-town-roots-and-identity",
+  "societal-pressure",
   "street-loyalty-and-danger",
-  "wealth-and-luxury",
-  "women-power"
+  "summer-as-metaphor",
+  "surrender-and-sacrifice",
+  "toxic-relationship",
+  "urgency-and-time-pressure",
+  "violence-and-street-danger",
+  "vulnerability-and-betrayal",
+  "wealth-and-flexing",
+  "wealth-and-material-success",
+  "women-power",
+  "youth-and-coming-up"
  ]
 }

runs/02-cluster/meta.json

@@ -0,0 +1,33 @@
+{
+ "script_version": "cluster_keywords.py 1.0.0",
+ "source_runs": [
+  "runs/01-tag/run1",
+  "runs/01-tag/run2"
+ ],
+ "source_records": [
+  882,
+  882
+ ],
+ "embedding": {
+  "model": "sentence-transformers/all-mpnet-base-v2",
+  "revision": "e8c3b32edf5434bc2275fc9bab85f82640a19130"
+ },
+ "clustering": {
+  "algorithm": "AgglomerativeClustering",
+  "linkage": "ward",
+  "metric": "euclidean",
+  "clusters": 100
+ },
+ "extra_keywords": [
+  "women-power"
+ ],
+ "keyword_count": 5999,
+ "versions": {
+  "python": "3.14.6",
+  "torch": "2.13.0+cpu",
+  "transformers": "5.14.1",
+  "sentence-transformers": "5.6.1",
+  "scikit-learn": "1.9.0",
+  "numpy": "2.5.1"
+ }
+}

runs/02-cluster/result-keywords.txt

@@ -1,50 +1,100 @@
-abandonment-and-solitude
+addiction-and-obsession
 alcohol-and-substance-abuse
+ambition-and-self-determination
 attraction-and-admiration
-betrayal-and-mistrust
-boastful-self-confidence
-breakup-and-reconciliation
+authenticity-and-honesty
+avoiding-commitment
+betrayal-and-cheating
+carefree-escapism
 celebration-and-partying
-contentment-and-joy
+cherishing-the-present-moment
+clinging-to-love
+community-and-togetherness
+confidence-and-self-assurance
+criminal-secrecy
 dancing-and-movement
-defiance-and-confrontation
-denial-and-pretense
-designer-fashion-flexing
-devotion-and-sacrifice
-embracing-true-identity
+deception-and-distrust
+defiance-and-rebellion
+denial-and-self-deception
+desire-and-lust
+devotion
+disappointment-and-failure
+dominance-and-power
+dream-and-awakening
+ex-moving-on
+faith-and-spirituality
+false-friendship
 fame-and-status
+family-and-distance
 family-and-fatherhood
-fantasy-and-imagination
-fear-of-vulnerability
+fear-of-losing-love
 female-empowerment
 flirtation-and-seduction
 freedom-and-escape
+gratitude-and-appreciation
+guilt-and-apology
+haters-and-critics
 heartbreak-and-grief
 hidden-emotional-struggle
-hope-and-disillusionment
+hope-and-optimism
 human-connection
-hustle-and-self-made-success
+hustle-and-money
+idealized-beauty
+impermanence
 independence-and-self-reliance
+individual-identity
+inner-mental-turmoil
 intimacy-and-connection
-jealousy-and-resentment-from-others
-longing-and-desire
+jealousy-and-bitterness
+legacy-and-family
+lingering-memories
+loneliness-and-isolation
+longing-and-loss
 longing-for-clarity
+longing-for-reconciliation
 love-as-fleeting-and-transient
-loyalty-and-commitment
+loyalty-and-relationships
+luxury-and-fashion
+mortality-and-death
+music-as-entertainment
 mutual-individuality
-nostalgia-and-memory
-obsession-and-madness
+nighttime-revelry
+non-judgment-and-acceptance
+nostalgia-and-longing
+overcoming-hardship
 partner-inadequacy
 partying-and-nightlife
+past-trauma-and-healing
 personal-growth-and-change
+possessiveness-and-exclusivity
+predatory-threat
+reassurance-and-comfort
+reassurance-and-protection
 reggaeton-culture
-regret-and-guilt
+regret-and-reflection
+rejection-of-women
+relationship-breakup
+relationship-complications
 relentless-ambition
-resilience-through-hardship
+reluctance-to-let-go
+reputation-and-gossip
+resilience-and-strength
+rising-to-success
 rivalry-and-superiority
 romantic-pursuit
-secrecy-and-paranoia
+self-confidence-and-braggadocio
 self-worth-and-insecurity
+sensual-desire
+showing-off-and-impressing
 small-town-roots-and-identity
+societal-pressure
 street-loyalty-and-danger
-wealth-and-luxury
+summer-as-metaphor
+surrender-and-sacrifice
+toxic-relationship
+urgency-and-time-pressure
+violence-and-street-danger
+vulnerability-and-betrayal
+wealth-and-flexing
+wealth-and-material-success
+youth-and-coming-up