ãããã¯ãŒã¯åæã«ãããã³ãã¥ããã£æ€åºãæ¢ãïŒæè¡ãæ¥çå¥å¿çšäŸãå°æ¥ã®ãã¬ã³ããè€éãªãããã¯ãŒã¯å ã®ã³ãã¥ããã£ãç¹å®ã»åæããæ¹æ³ãçè§£ããŸãã
ãããã¯ãŒã¯åæïŒã³ãã¥ããã£æ€åºã«ããã€ã³ãµã€ãã®çºèŠ
仿¥ã®çžäºæ¥ç¶ãããäžçã§ã¯ãè€éãªã·ã¹ãã å ã®è€éãªé¢ä¿æ§ãçè§£ããããšãäžå¯æ¬ ã§ãããããã¯ãŒã¯åæã¯ãããã®ã·ã¹ãã ãåæããããã®ããŒã«ãšæè¡ãæäŸããã³ãã¥ããã£æ€åºã¯é ããæ§é ãšäŸ¡å€ããã€ã³ãµã€ããæããã«ããããã®åŒ·åãªã¢ãããŒããšããŠéç«ã£ãŠããŸãããã®å æ¬çãªã¬ã€ãã§ã¯ãã³ãã¥ããã£æ€åºã®åºæ¬æŠå¿µããã®å€æ§ãªå¿çšããããŠãã®åºæ¿çãªåéã®é²åãã屿ã«ã€ããŠæ¢ããŸãã
ã³ãã¥ããã£æ€åºãšã¯ïŒ
æ žå¿ãšããŠãã³ãã¥ããã£æ€åºã¯ããããã¯ãŒã¯å ã®ããŒãïŒãŸãã¯é ç¹ïŒã®ã°ã«ãŒãã§ããããã¯ãŒã¯ã®ä»ã®éšåãããäºãã«å¯æ¥ã«æ¥ç¶ãããŠãããã®ãç¹å®ããããšãç®çãšããŸãããããã®ã°ã«ãŒãã¯ããã°ãã°ã³ãã¥ããã£ãã¯ã©ã¹ã¿ãŒããŸãã¯ã¢ãžã¥ãŒã«ãšåŒã°ããŸããç®æšã¯ããããã¯ãŒã¯ããããã®æå³ã®ããéšåæ§é ã«åå²ããä»ã§ã¯èŠéããããã¡ãªãã¿ãŒã³ãé¢ä¿æ§ãæããã«ããããšã§ãã
ãœãŒã·ã£ã«ãããã¯ãŒã¯ãæ³åããŠã¿ãŠãã ãããã³ãã¥ããã£æ€åºã¯ãå人ãååããŸãã¯å ±éã®èå³ãæã€å人ã®ã°ã«ãŒããç¹å®ããã®ã«åœ¹ç«ã¡ãŸããåæ§ã«ãã¿ã³ãã¯è³ªéçžäºäœçšã衚ãçç©åŠçãããã¯ãŒã¯ã§ã¯ãæ©èœçãªã¢ãžã¥ãŒã«ãçµè·¯ãæããã«ããããšãã§ããŸãããµãã©ã€ãã§ãŒã³ãããã¯ãŒã¯ã§ã¯ã匷åãªååŒé¢ä¿ã§çµã°ããããžãã¹ã®ã¯ã©ã¹ã¿ãŒãèŠã€ããã®ã«åœ¹ç«ã¡ãŸãããã®å¿çšã¯åºç¯ãã€å€æ§ã§ãã
äž»èŠãªæŠå¿µ
- ããŒãïŒé ç¹ïŒïŒ ãããã¯ãŒã¯å ã®åã ã®ãšã³ãã£ãã£ïŒäŸïŒãœãŒã·ã£ã«ãããã¯ãŒã¯ã«ããã人ã ãçç©åŠçãããã¯ãŒã¯ã«ãããã¿ã³ãã¯è³ªãã€ã³ã¿ãŒãããäžã®ãŠã§ããµã€ãïŒã
- ãšããžïŒãªã³ã¯ïŒïŒ ããŒãéã®æ¥ç¶ãŸãã¯é¢ä¿ïŒäŸïŒãœãŒã·ã£ã«ãããã¯ãŒã¯ã«ãããåæ ãã¿ã³ãã¯è³ªéçžäºäœçšããŠã§ããµã€ãéã®ãã€ããŒãªã³ã¯ïŒã
- ãããã¯ãŒã¯ïŒã°ã©ãïŒïŒ ç 究察象ã®ã·ã¹ãã ã衚ãããŒããšãšããžã®éåã
- ã³ãã¥ããã£ïŒ ãããã¯ãŒã¯ã®ä»ã®éšåãããäºãã«å¯æ¥ã«æ¥ç¶ãããŠããããŒãã®ã°ã«ãŒãã
- ã¢ãžã¥ã©ãªãã£ïŒ ã³ãã¥ããã£æ§é ã®å質ãå®éåããææšãã¢ãžã¥ã©ãªãã£ãé«ãã»ã©ãããè¯ãã³ãã¥ããã£åå²ã瀺ããŸãã
ãªãã³ãã¥ããã£æ€åºã¯éèŠãªã®ãïŒ
ã³ãã¥ããã£æ€åºã¯ãããŸããŸãªé åã§ããã€ãã®äž»èŠãªå©ç¹ãæäŸããŸãïŒ
- ãããã¯ãŒã¯æ§é ã®çè§£ïŒ è€éãªãããã¯ãŒã¯ã®æ ¹åºã«ããçµç¹ãæããã«ããã·ã¹ãã ã®ç°ãªãéšåãã©ã®ããã«çžäºäœçšãã圱é¿ãåããã«ã€ããŠã®ã€ã³ãµã€ããæäŸããŸãã
- äž»èŠãªãã¬ã€ã€ãŒã®ç¹å®ïŒ ã³ãã¥ããã£ã®æå±ãæ¥ç¶ãåæããããšã§ãç¹å®ã®ã³ãã¥ããã£å ã®åœ±é¿åã®ããå人ããšã³ãã£ãã£ãç¹å®ããã®ã«åœ¹ç«ã¡ãŸãã
- æšèŠã®æ¹åïŒ æšèŠã·ã¹ãã ã«ãããŠãã³ãã¥ããã£æå±ãèæ ®ããããšã§ãææ¡ã®ç²ŸåºŠãšé¢é£æ§ãé«ããããšãã§ããŸããäŸãã°ããŠãŒã¶ãŒãç¹å®ãããã³ãã¥ããã£å ã§äººæ°ã®ãã補åããµãŒãã¹ãæšèŠãããªã©ã§ãã
- ç°åžžãšäžæ£ã®æ€åºïŒ ã³ãã¥ããã£å ã®éåžžãšã¯ç°ãªãæ¥ç¶ãã¿ãŒã³ã¯ãç°åžžãäžæ£è¡çºã®å åãšãªãå¯èœæ§ããããŸãã
- å°æ¥ã®è¡åäºæž¬ïŒ ã³ãã¥ããã£æ§é ãçè§£ããããšã¯ãæ å ±ã圱é¿ããããã¯ãŒã¯ãéããŠã©ã®ããã«åºããããäºæž¬ããã®ã«åœ¹ç«ã¡ãŸãã
äžè¬çãªã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã
ã³ãã¥ããã£æ€åºã®ããã«ã¯æ°å€ãã®ã¢ã«ãŽãªãºã ãéçºãããŠãããããããã«é·æãšçæããããŸãã以äžã«ãæãåºã䜿çšãããŠããã¢ãããŒãã®ããã€ããæããŸãïŒ
1. ã«ãŒãã³æ³ (Louvain Algorithm)
ã«ãŒãã³æ³ã¯ãã¢ãžã¥ã©ãªãã£ãæå€§åããããšãç®çãšãã貪欲ãªéå±€çã¢ã«ãŽãªãºã ã§ãããããã¯ãŒã¯ã®ã¢ãžã¥ã©ãªãã£ã屿çãªæå€§å€ã«éãããŸã§ãããŒããã³ãã¥ããã£éã§ç¹°ãè¿ãç§»åãããŸãããã®ã¢ã«ãŽãªãºã ã¯é床ãšã¹ã±ãŒã©ããªãã£ã§ç¥ãããŠãããå€§èŠæš¡ãªãããã¯ãŒã¯ã«é©ããŠããŸãããœãŒã·ã£ã«ãããã¯ãŒã¯åæããã®ä»ã®ã¢ããªã±ãŒã·ã§ã³ã§åºã䜿çšãããŠããŸãã
äŸïŒ å€§èŠæš¡ãªãªã³ã©ã€ã³ãã©ãŒã©ã ãæ³åããŠãã ãããã«ãŒãã³æ³ã䜿çšããŠãã©ãŒã©ã å ã®ç°ãªããããã¯ããŒã¹ã®ã³ãã¥ããã£ãç¹å®ããã¢ãã¬ãŒã¿ãŒããŠãŒã¶ãŒã®èå³ãããããçè§£ããã³ã³ãã³ããé©å®èª¿æŽããããšãã§ããŸãã
2. ã¬ãŒãã³ã»ãã¥ãŒãã³æ³ (Girvan-Newman Algorithm / åªä»äžå¿æ§)
ã¬ãŒãã³ã»ãã¥ãŒãã³æ³ã¯ãåªä»äžå¿æ§ã¢ã«ãŽãªãºã ãšããŠãç¥ãããåå²çãªã¢ãããŒããåããŸããåªä»äžå¿æ§ïŒãã¹ãŠã®ããŒããã¢éã®æççµè·¯ã®ãã¡ããã®ãšããžãééãããã®ã®æ°ïŒãæãé«ããšããžãç¹°ãè¿ãåé€ãããããã¯ãŒã¯ãéé£çµãªã³ã³ããŒãã³ãïŒããããã³ãã¥ããã£ãšèŠãªãããïŒã«åè§£ããããŸã§ç¶ããŸããæŠå¿µçã«ã¯åçŽã§ããããã®ã¢ã«ãŽãªãºã ã¯å€§èŠæš¡ãªãããã¯ãŒã¯ã§ã¯èšç®ã³ã¹ããé«ããªãå¯èœæ§ããããŸãã
äŸïŒ 亀éãããã¯ãŒã¯ã«ãããŠãã¬ãŒãã³ã»ãã¥ãŒãã³æ³ã¯ãåé€ããããšç¹å®ã®å°åãã³ãã¥ããã£ãå€ç«ãããå¯èœæ§ã®ããéèŠãªæ¥ç¶ãæ©ãç¹å®ããããšãã§ããŸãã
3. ã©ãã«äŒææ³ (Label Propagation Algorithm)
ã©ãã«äŒææ³ã¯ãåããŒãã«äžæã®ã©ãã«ãå²ãåœãŠãã·ã³ãã«ã§å¹ççãªã¢ã«ãŽãªãºã ã§ããããŒãã¯ã飿¥ããŒãã®äžã§æãé »ç¹ã«åºçŸããã©ãã«ã«äžèŽããããã«ãèªèº«ã®ã©ãã«ãç¹°ãè¿ãæŽæ°ããŸãããã®ããã»ã¹ã¯ãåããŒãã飿¥ããŒãã®å€§å€æ°ãšåãã©ãã«ãæã€ãŸã§ç¶ããŸããéåžžã«é«éã§å€§èŠæš¡ãªãããã¯ãŒã¯ã«é©ããŠããŸãããåæã®ã©ãã«å²ãåœãŠã«ææãªå ŽåããããŸãã
äŸïŒ ç ç©¶è ãšãã®åºçç©ã®ãããã¯ãŒã¯ãèããŠã¿ãŸããããã©ãã«äŒææ³ã䜿çšãããšãåºçç©ã«ãããåŒçšãã¿ãŒã³ã«åºã¥ããŠãé¢é£ãããããã¯ã«åãçµãã§ããç ç©¶è ã®ã³ãã¥ããã£ãç¹å®ã§ããŸãã
4. ã©ã€ãã³æ³ (Leiden Algorithm)
ã©ã€ãã³æ³ã¯ã«ãŒãã³æ³ãæ¹è¯ãããã®ã§ãããæ¥ç¶æ§ã®äœãã³ãã¥ããã£ãçæããåŸåããããšãã£ãã«ãŒãã³æ³ã®æ¬ ç¹ã«å¯ŸåŠããŠããŸããç¹å®ãããåã³ãã¥ããã£ãé£çµæåã§ããããšãä¿èšŒããããåªããçè«çåºç€ãæäŸããŸãããã®ç²ŸåºŠãšå ç¢æ§ããããŸããŸã人æ°ãé«ãŸã£ãŠããŸãã
äŸïŒ å€§èŠæš¡ãªéºäŒåå¶åŸ¡ãããã¯ãŒã¯ã«ãããŠãã©ã€ãã³æ³ã¯ã«ãŒãã³æ³ãšæ¯èŒããŠãããå®å®ãæç¢ºã«å®çŸ©ãããæ©èœçã¢ãžã¥ãŒã«ãç¹å®ã§ããå¯èœæ§ããããéºäŒåçžäºäœçšã®ããè¯ãçè§£ã«ã€ãªãããŸãã
5. ã€ã³ãã©ãããæ³ (Infomap Algorithm)
ã€ã³ãã©ãããæ³ã¯ããããã¯ãŒã¯äžã®ã©ã³ãã ãŠã©ãŒã«ãŒã®åãã®èšè¿°é·ãæå°åãããšããåçã«åºã¥ããŠããŸããæ å ±çè«ã䜿çšããŠããŠã©ãŒã«ãŒã®çµè·¯ãèšè¿°ããããã«å¿ èŠãªæ å ±éãæå°åããã³ãã¥ããã£ãèŠã€ããŸããç¹ã«æåãããã¯ãŒã¯ããããŒã®ãããããã¯ãŒã¯ã«å¹æçã§ãã
äŸïŒ ã€ã³ã¿ãŒãããäžã®æ å ±ã®æµãã衚ããããã¯ãŒã¯ãæ³åããŠãã ãããã€ã³ãã©ãããæ³ã¯ãé »ç¹ã«äžç·ã«èšªåããããŠã§ããµã€ãã®ã³ãã¥ããã£ãç¹å®ããæ å ±æ¶è²»ã®ãã¿ãŒã³ãæããã«ããããšãã§ããŸãã
6. ã¹ãã¯ãã«ã¯ã©ã¹ã¿ãªã³ã°
ã¹ãã¯ãã«ã¯ã©ã¹ã¿ãªã³ã°ã¯ããããã¯ãŒã¯ã®é£æ¥è¡åãŸãã¯ã©ãã©ã·ã¢ã³è¡åã®åºæå€ãšåºæãã¯ãã«ã䜿çšããŠãã¯ã©ã¹ã¿ãªã³ã°ã®åã«ããŒã¿ã®æ¬¡å ãåæžããŸããç¹ã«éåžãªã¯ã©ã¹ã¿ãŒåœ¢ç¶ã«å¯ŸããŠãåŸæ¥ã®ã¯ã©ã¹ã¿ãªã³ã°ã¢ã«ãŽãªãºã ããã粟床ãé«ãããšãå€ãã§ããããããéåžžã«å€§èŠæš¡ãªãããã¯ãŒã¯ã§ã¯èšç®ã³ã¹ããé«ããªãå¯èœæ§ããããŸãã
äŸïŒ ç»åã»ã°ã¡ã³ããŒã·ã§ã³ã«ãããŠãã¹ãã¯ãã«ã¯ã©ã¹ã¿ãªã³ã°ã䜿çšããŠããã¯ã»ã«ãé¡äŒŒæ§ã«åºã¥ããŠç°ãªãé åã«ã°ã«ãŒãåããåããªããžã§ã¯ãã«å±ãããã¯ã»ã«ã®ã³ãã¥ããã£ã广çã«ç¹å®ããããšãã§ããŸãã
ã³ãã¥ããã£æ§é ã®è©äŸ¡
ã³ãã¥ããã£æ§é ãç¹å®ããããããã®å質ãè©äŸ¡ããããšãéèŠã§ããã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ã®æå¹æ§ãè©äŸ¡ããããã«ãããã€ãã®ææšã䜿çšã§ããŸãïŒ
- ã¢ãžã¥ã©ãªã㣠(Q): åè¿°ã®éããã¢ãžã¥ã©ãªãã£ã¯ã³ãã¥ããã£å ã®æ¥ç¶å¯åºŠãšã³ãã¥ããã£éã®æ¥ç¶å¯åºŠãæ¯èŒããŠå®éåããŸããã¢ãžã¥ã©ãªãã£ã¹ã³ã¢ãé«ãã»ã©ãããè¯ãã³ãã¥ããã£åå²ã瀺ããŸãã
- æ£èŠåçžäºæ å ±é (NMI): NMIã¯ã2ã€ã®ç°ãªãã³ãã¥ããã£æ§é éã®é¡äŒŒæ§ã枬å®ããŸããç°ãªãã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ã®çµæãæ¯èŒããããäºæž¬ãããã³ãã¥ããã£æ§é ãæ£è§£ããŒã¿ïŒããããã°ïŒãšæ¯èŒãããããããã«ãã䜿çšãããŸãã
- 調æŽã©ã³ãææ° (ARI): ARIã¯ãå¶ç¶ã®äžèŽã®å¯èœæ§ãèæ ®ã«å ¥ããŠã2ã€ã®ç°ãªãã¯ã©ã¹ã¿ãªã³ã°ãæ¯èŒããããã®ãã1ã€ã®ææšã§ãã
- ã³ã³ãã¯ã¿ã³ã¹ïŒ ã³ã³ãã¯ã¿ã³ã¹ã¯ãã³ãã¥ããã£å ã®ç·ãšããžæ°ã«å¯ŸããŠãã³ãã¥ããã£ããåºãŠãããšããžã®å²åãæž¬å®ããŸããã³ã³ãã¯ã¿ã³ã¹ãäœãã»ã©ãããçµæåã®é«ãã³ãã¥ããã£ã瀺ããŸãã
- ã«ãã¬ããžïŒ ã«ãã¬ããžã¯ãç¹å®ãããã³ãã¥ããã£å ã«å«ãŸãããšããžã®å²åãæž¬å®ããŸããã«ãã¬ããžãé«ãã»ã©ãã³ãã¥ããã£æ§é ããããã¯ãŒã¯ã®æ¥ç¶ã®å€§éšåãæããŠããããšã瀺ããŸãã
æ¥çãè¶ ããã³ãã¥ããã£æ€åºã®å¿çš
ã³ãã¥ããã£æ€åºã¯ãåºç¯ãªæ¥çãåéã§å¿çšãããŠããŸãïŒ
1. ãœãŒã·ã£ã«ãããã¯ãŒã¯åæ
ããã¯æãé¡èãªå¿çšã®1ã€ã§ããã³ãã¥ããã£æ€åºã¯ãFacebookãTwitterãLinkedInãªã©ã®ãã©ãããã©ãŒã äžã§ãå人ãååããŸãã¯å ±éã®èå³ãæã€å人ã®ã°ã«ãŒããç¹å®ããããã«äœ¿çšãããŸãããã®æ å ±ã¯ãã¿ãŒã²ããåºåãããŒãœãã©ã€ãºãããæšèŠã瀟äŒåæ ã®çè§£ã«å©çšã§ããŸãã
äŸïŒ ç¹å®ã®ãããã¯ã«èå³ãæã€ãŠãŒã¶ãŒã®ã³ãã¥ããã£ãç¹å®ãããã©ãããã©ãŒã ãããé¢é£æ§ã®é«ãã³ã³ãã³ããæšèŠãæäŸã§ããããã«ããŸãã
2. ãã€ãªã€ã³ãã©ããã£ã¯ã¹
ãã€ãªã€ã³ãã©ããã£ã¯ã¹ã§ã¯ãã³ãã¥ããã£æ€åºã¯ãã¿ã³ãã¯è³ªéçžäºäœçšãããã¯ãŒã¯ãéºäŒåå¶åŸ¡ãããã¯ãŒã¯ã代è¬ãããã¯ãŒã¯ã«ãããæ©èœçã¢ãžã¥ãŒã«ãç¹å®ããããã«äœ¿çšãããŸãããããã®ã¢ãžã¥ãŒã«ã¯ãç¹å®ã®æ©èœãå®è¡ããçµè·¯ãè€åäœããŸãã¯ãã®ä»ã®çç©åŠçåäœã衚ãããšããããŸãã
äŸïŒ ã¿ã³ãã¯è³ªéçžäºäœçšãããã¯ãŒã¯å ã®ã¿ã³ãã¯è³ªè€åäœãç¹å®ããç ç©¶è ã现èããã»ã¹ãéè¡ããããã«ã¿ã³ãã¯è³ªãã©ã®ããã«çžäºäœçšããããçè§£ããã®ã«åœ¹ç«ã¡ãŸãã
3. éä¿¡ãããã¯ãŒã¯
ã³ãã¥ããã£æ€åºã¯ãéä¿¡ãããã¯ãŒã¯ã®æ§é ãåæããäºãã«é »ç¹ã«éä¿¡ãããŠãŒã¶ãŒã®ã¯ã©ã¹ã¿ãŒãç¹å®ããããã«äœ¿çšã§ããŸãããã®æ å ±ã¯ããããã¯ãŒã¯ã®æé©åããã©ãã£ãã¯ç®¡çãäžæ£æ€åºã«å©çšã§ããŸãã
äŸïŒ é »ç¹ã«äºãã«é話ããæºåž¯é»è©±ãŠãŒã¶ãŒã®ã³ãã¥ããã£ãç¹å®ããéä¿¡äŒç€Ÿããããã¯ãŒã¯ãªãœãŒã¹ãæé©åããã¿ãŒã²ãããçµã£ããµãŒãã¹ãæäŸã§ããããã«ããŸãã
4. 亀éãããã¯ãŒã¯
亀éãããã¯ãŒã¯ã§ã¯ãã³ãã¥ããã£æ€åºã¯äº€éãªã³ã¯ã«ãã£ãŠåŒ·ãçµã°ããŠããéœåžãå°åã®ã¯ã©ã¹ã¿ãŒãç¹å®ã§ããŸãããã®æ å ±ã¯ãéœåžèšç»ã亀éã€ã³ãã©éçºãç·æ¥å¯Ÿå¿ã«å©çšã§ããŸãã
äŸïŒ é »ç¹ãªãã©ã€ãã§çµã°ããŠããéœåžã®ã³ãã¥ããã£ãç¹å®ãã亀éèšç»è ããã©ã€ãã¹ã±ãžã¥ãŒã«ãæé©åããæ¥ç¶æ§ãåäžãããããšãã§ããŸãã
5. éèãšäžæ£æ€åº
ã³ãã¥ããã£æ€åºã¯ãäžæ£è¡çºã«é¢äžããŠããå人ãçµç¹ã®ã°ã«ãŒããç¹å®ããããã«äœ¿çšã§ããŸããååŒãããã¯ãŒã¯ãåæããéåžžãšã¯ç°ãªãæ¥ç¶ãã¿ãŒã³ãç¹å®ããããšã§ããããŒãã³ããªã³ã°ãã€ã³ãµã€ããŒååŒããã®ä»ã®åœ¢æ ã®éèè©æ¬ºã®æ€åºã«åœ¹ç«ã¡ãŸãã
äŸïŒ äžå¯©ãªååŒã«é¢äžããŠããã¢ã«ãŠã³ãã®ã°ã«ãŒããç¹å®ããäžæ£æ€åºã¢ããªã¹ãã«ãããããªã調æ»ã®ããã«ãã©ã°ãç«ãŠãŸãã
6. æ å ±æ€çŽ¢ãšæšèŠã·ã¹ãã
ã³ãã¥ããã£æ€åºã¯ããŠãŒã¶ãŒãšã¢ã€ãã ã®ã³ãã¥ããã£æå±ãèæ ®ããããšã§ãæšèŠã®ç²ŸåºŠãšé¢é£æ§ãåäžãããããšãã§ããŸããäŸãã°ãæšèŠã·ã¹ãã ã¯ããŠãŒã¶ãŒãç¹å®ãããã³ãã¥ããã£å ã§äººæ°ã®ããã¢ã€ãã ãææ¡ããããšããããŸãã
äŸïŒ ç¹å®ãããæ ç»éè³ã³ãã¥ããã£å ã®ä»ã®ãŠãŒã¶ãŒã®å¥œã¿ã«åºã¥ããŠããŠãŒã¶ãŒã«æ ç»ãæšèŠããŸãã
7. ãµãã©ã€ãã§ãŒã³åæ
ã³ãã¥ããã£æ€åºã¯ããµãã©ã€ãã§ãŒã³ãããã¯ãŒã¯ã®æ§é ãåæããååŒé¢ä¿ã«ãã£ãŠåŒ·ãçµã°ããŠããããžãã¹ã®ã¯ã©ã¹ã¿ãŒãç¹å®ããããã«äœ¿çšã§ããŸãããã®æ å ±ã¯ããªã¹ã¯ç®¡çããµãã©ã€ãã§ãŒã³ã®æé©åãæœåšçãªæ··ä¹±ã®ç¹å®ã«å©çšã§ããŸãã
äŸïŒ äºãã«åŒ·ãäŸåããŠãããµãã©ã€ã€ãŒãšè£œé æ¥è ã®ã³ãã¥ããã£ãç¹å®ããäŒæ¥ããµãã©ã€ãã§ãŒã³ã®æ··ä¹±ãªã¹ã¯ã軜æžã§ããããã«ããŸãã
8. ãªã³ã©ã€ã³ã³ãã¥ããã£
ãªã³ã©ã€ã³ãã©ãŒã©ã ããœãŒã·ã£ã«ã¡ãã£ã¢ã°ã«ãŒãããã®ä»ã®ãªã³ã©ã€ã³ã³ãã¥ããã£ãåæããŠããã®æ§é ãçè§£ãã圱é¿åã®ããã¡ã³ããŒãç¹å®ããæ°ããªãã¬ã³ããæ€åºããŸãã
äŸïŒ ç¹å®ã®ãããã¯ã«ã€ããŠæŽ»çºã«è°è«ããŠãããŠãŒã¶ãŒã®ã³ãã¥ããã£ãç¹å®ãããã©ãããã©ãŒã 管çè ãã³ã³ãã³ããã¢ãã¬ãŒãããçç£çãªè°è«ãä¿é²ã§ããããã«ããŸãã
ã³ãã¥ããã£æ€åºã®ããã®ããŒã«ãšæè¡
ã³ãã¥ããã£æ€åºãå®è¡ããããã«ãããã€ãã®ãœãããŠã§ã¢ããŒã«ãšã©ã€ãã©ãªãå©çšå¯èœã§ãïŒ
- NetworkX (Python): ãããã¯ãŒã¯ã®äœæãæäœãåæã®ããã®äººæ°ã®Pythonã©ã€ãã©ãªãããã€ãã®ã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ã®å®è£ ãå«ãŸããŠããŸãã
- igraph (R, Python, C++): ãããã¯ãŒã¯åæã§åºã䜿çšãããŠãããã1ã€ã®ã©ã€ãã©ãªã§ãåºç¯ãªã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ããã®ä»ã®ãããã¯ãŒã¯åæããŒã«ãæäŸããŠããŸãã
- Gephi: ã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ãå«ãããªãŒãã³ãœãŒã¹ã®ã°ã©ãå¯èŠåã»åæãœãããŠã§ã¢ããã±ãŒãžã
- ã°ã©ãããŒã¿ããŒã¹ãã©ãããã©ãŒã (Neo4j, Amazon Neptune): ã°ã©ãããŒã¿ããŒã¹ã¯ãããã¯ãŒã¯ããŒã¿ã®å¹ççãªä¿åãšæ€çŽ¢ãæäŸããå€§èŠæš¡ãããã¯ãŒã¯ã§ã®ã³ãã¥ããã£æ€åºã«é©ããŠããŸããå€ãã®ã°ã©ãããŒã¿ããŒã¹ã«ã¯ãçµã¿èŸŒã¿ã®ã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ãå€éšã©ã€ãã©ãªãšã®çµ±åæ©èœãå«ãŸããŠããŸãã
- SNAP (Stanford Network Analysis Platform): å€§èŠæš¡ãããã¯ãŒã¯åæã®ããã®æ±çšçãªé«æ§èœã·ã¹ãã ãå€§èŠæš¡ã°ã©ãåãã«æé©åãããããã€ãã®ã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ãæäŸããŠããŸãã
課é¡ãšä»åŸã®æ¹åæ§
ã³ãã¥ããã£æ€åºã«ããã倧ããªé²æ©ã«ãããããããããã€ãã®èª²é¡ãæ®ã£ãŠããŸãïŒ
- ã¹ã±ãŒã©ããªãã£ïŒ å€ãã®ã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ã¯ãæ°çŸäžãŸãã¯æ°ååã®ããŒããšãšããžãæã€éåžžã«å€§èŠæš¡ãªãããã¯ãŒã¯ãžã®ã¹ã±ãŒãªã³ã°ã«èŠåŽããŠããŸããããå¹ççã§ã¹ã±ãŒã©ãã«ãªã¢ã«ãŽãªãºã ã®éçºã¯ãçŸåšé²è¡äžã®ç ç©¶åéã§ãã
- éè€ããã³ãã¥ããã£ïŒ å€ãã®å®äžçã®ãããã¯ãŒã¯ã§ã¯ãããŒãã¯è€æ°ã®ã³ãã¥ããã£ã«åæã«å±ããããšããããŸããéè€ããã³ãã¥ããã£ãæ£ç¢ºã«æ€åºã§ããã¢ã«ãŽãªãºã ã®éçºã¯ãå°é£ãªåé¡ã§ãã
- åçãããã¯ãŒã¯ïŒ ãããã¯ãŒã¯ã¯æéãšãšãã«é²åããããŒãããšããžã远å ãŸãã¯åé€ãããŸããåçãããã¯ãŒã¯ã«ãããã³ãã¥ããã£æ§é ã远跡ã§ããã¢ã«ãŽãªãºã ã®éçºã¯ãéèŠãªç ç©¶åéã§ãã
- æåã»éã¿ä»ããããã¯ãŒã¯ã«ãããã³ãã¥ããã£æ€åºïŒ å€ãã®ã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ã¯ãç¡åã»éã¿ãªããããã¯ãŒã¯çšã«èšèšãããŠããŸãããããã®ã¢ã«ãŽãªãºã ãæåã»éã¿ä»ããããã¯ãŒã¯ã«å¯Ÿå¿ãããããšã¯ãçŸåšé²è¡äžã®èª²é¡ã§ãã
- æ£è§£ããŒã¿ã«ããæ€èšŒïŒ ã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ã®ç²ŸåºŠãè©äŸ¡ããããšã¯å°é£ãªå ŽåããããŸããç¹ã«ãæ£è§£ãšãªãã³ãã¥ããã£æ§é ãå©çšã§ããªãå Žåã¯ãªãããã§ããã³ãã¥ããã£æ€åºã®çµæãæ€èšŒããããã®ããè¯ãæ¹æ³ã®éçºã¯ãéèŠãªç ç©¶åéã§ãã
ã³ãã¥ããã£æ€åºã«ãããå°æ¥ã®ç ç©¶ã®æ¹åæ§ã¯æ¬¡ã®ãšããã§ãïŒ
- ããã¹ã±ãŒã©ãã«ã§å¹ççãªã¢ã«ãŽãªãºã ã®éçºã
- éè€ããã³ãã¥ããã£ãæ€åºããããã®ã¢ã«ãŽãªãºã ã®éçºã
- åçãããã¯ãŒã¯ã«ãããã³ãã¥ããã£æ§é ã远跡ããããã®ã¢ã«ãŽãªãºã ã®éçºã
- æåã»éã¿ä»ããããã¯ãŒã¯ã«ãããã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ã®éçºã
- ã³ãã¥ããã£æ€åºçµæãæ€èšŒããããã®ããè¯ãæ¹æ³ã®éçºã
- ã³ãã¥ããã£æ€åºãšä»ã®æ©æ¢°åŠç¿æè¡ãšã®çµ±åã
- ã³ãã¥ããã£æ€åºãæ°èŠããã³æ°èã®ã¢ããªã±ãŒã·ã§ã³ã«é©çšããããšã
çµè«
ã³ãã¥ããã£æ€åºã¯ãè€éãªãããã¯ãŒã¯å ã®é ããæ§é ãšäŸ¡å€ããã€ã³ãµã€ããæããã«ããããã®åŒ·åãªæè¡ã§ãããœãŒã·ã£ã«ãããã¯ãŒã¯åæãããã€ãªã€ã³ãã©ããã£ã¯ã¹ãéèã«è³ããŸã§ãæ¥çãè¶ ãã倿§ãªå¿çšã¯ã仿¥ã®ããŒã¿é§åå瀟äŒã«ããããã®éèŠæ§ãæµ®ã圫ãã«ããŠããŸãããããã¯ãŒã¯ã®èŠæš¡ãšè€éããå¢ãç¶ããäžã§ãããå¹ççã§ãæ£ç¢ºã§ãã¹ã±ãŒã©ãã«ãªã³ãã¥ããã£æ€åºã¢ã«ãŽãªãºã ã®éçºã¯ããã®æœåšèœåãæå€§éã«åŒãåºãããã«äžå¯æ¬ ãšãªãã§ããããã³ãã¥ããã£æ€åºã®ååãšæè¡ãçè§£ããããšã§ãããŸããŸãªåéã®å°éå®¶ã¯ãç 究察象ã®ã·ã¹ãã ãããæ·±ãçè§£ããããæ å ±ã«åºã¥ããæææ±ºå®ãè¡ãããšãã§ããŸãã