Xây dựng hệ thống tư vấn trên các trang facebook fanpage

TRÊN CÁC TRANG FACEBOOK FANPAGE Chuyên ngành : 60.48.01 11 2013 -HCM HCM ngày 25 tháng 12 Tr 2013 ng i h c Bách Khoa, HQG Tp Thành ph n H i ng nh giá lu n v n th c s g m: (Ghi c a H i ng ch m b o v lu n v n th c s) Xác nh n c a Ch t ch H i ng chuyên ngành sau lu n v n ã h giá LV c s a ch a (n u có) KHOA ii lý NAM - - KHOA MSHV: 11070441 04/06/1979 KHÁNH HÒA : 60.48.01 Chuyên ngành: I FANPAGE II P III : 21/01/2013 IV : 22/11/2013 V : Tp HCM, ng iii 20 iv m v ABSTRACT In recent years, social network has changed completely the connection method among netizens and become an essential part of the daily lives of billions of people around the world The continuous growth of social network has made the process of finding useful information become difficult, and it also motivates the research of data mining on social network environment more powerful One of the research directions in this field is data mining on social networks to advise users with useful, quick and accurate data Within the scope of this thesis, we researched on classic methods in data mining, assessment methods, user classification in the social network After that, we proposed a method of combining between fanpage content for advising people to suitable fanpages vi vii iv v ABSTRACT vi vii viii x xi 12 1.1 12 1.2 13 1.3 14 16 2.1 16 2.2 17 2.2.1 17 2.2.2 18 2.3 2.3.1 19 19 2.3.2 -means 19 2.4 2.5 3.1 3.1.1 25 25 27 27 Vector space model 27 3.1.2 28 3.1.3 28 3.2 3.2.1 29 30 3.2.2 31 3.3 3.4 32 Social Lens 33 4.1 4.2 4.3 4.3.1 35 35 36 37 37 4.3.2 40 viii 4.4 4.4.1 40 40 4.4.2 41 4.5 4.5.1 42 42 4.5.2 42 4.6 4.7 5.1 5.1.1 43 43 44 44 44 a 44 b 51 c 54 5.1.2 Ph 54 a 55 b 55 : 56 56 58 59 7: 65 65 66 68 69 69 72 75 76 6.1 6.2 6.3 7.1 7.2 ix D 39 -strength 39 x : Thínghi 64 t : T ng k t 7.1 K t lu n khai thác gom nhóm dung fanpage dùng Xây d ng ng d d li u facebook fanpage, ng d ng m fanpage m nh fanpage ng ng ho t nh fanpage có t n su t c p nh ng d ng ng th n x lý d li u, cung c p cho h th ng thông tin c phân tách n i dung (tên fanpage, s t like, n i dung fanpage, s Xây d ng thành ph n ti n x m nhi m vi c lo i b t d ng (stop-word), lo i b ti n t h u t (stemming) Phát tri n module xây d ng không gian t vector tr ng s tf-idf Xây d ng module phân c làm gi m khơng gian tìm ki m trình tìm ki m d li u Xây d cung c n ng d ng g i ý fanpage 65 : T ng k t Xây d ng ng d t khóa t 7.2 ph c v cho vi c nh p i dùng hi n th k t qu tìm ki m ng phát tri n Nghiên c u phát tri n thu n i dung c a hai hay nhi u fanpages, t ng g i ý xác i dùng Nghiên c u phát tri n chi tìm ki c c p nh t d li u nh c c p nh t g n d li u th t nh t Nghiên c u s d i dùng, ph c v cho vi c cung c p k t qu h pv it H th m b o d li u ng (k t qu tr v s phù ng tìm ki m) c xây d ng khuôn kh nghiên c u ch áp d ng cho fanpage s d ng ti ng 66 : T ng k t phát tri n m r ng c a nghiên c u x lý nhi u ngôn ng c bi t ti ng Vi t Trong khuôn kh c a nghiên c u này, x lý phân lo b n thành vector tr ng s tf- n u gây h n ch vi c x lý ngơn ng có t ghép X lý t câu truy v n ng m ng phát tri n m r ng t 67 n tài Tàiliu tham kh o [1] R A Baeza-Yates, C A Hurtado, and M Mendoza Improving search engines by query clustering JASIST , 58(12):1793 1804, 2007 [2] D Beeferman and A L Berger Agglomerative clustering of a search engine query log In Proceedings of the 6th ACM SIGKDD International Conference on Knowledge 416, Boston, MA, USA, 2000 [3] Elizabeth M Daly, Michael J Muller, Liang Gou, David R Millen Social Lens: Personalization Around User Defined Collections for Filtering Enterprise Message Streams Proceedings of the Fifth International Conference on Weblogs and Social Media, Barcelona, Catalonia, Spain, July 17-21, 2011 [4] L Li, S Otsuka, and M Kitsuregawa Query recommendation using largescale web access logs and web page archive In Proceedings of 19th International Conference on Database and Expert Systems Applications 34 141, Turin, Italy, 2008 [5] H Ma, H Yang, I King, and M R Lyu Learning latent semantic relations from click-through data for query suggestion In Proceedings of the 17th ACM Conference on pages 709 718, Napa Valley, California,USA, 2008 [6] Q Mei, D Zhou, and K W Church Query suggestion using hitting time InProceedings of the 17th ACM Conference on Information and Knowledge pages 469 478, Napa Valley, California, USA, 2008 [7] Guandong Xu, Web Mining and Social Networking Techniques and Applications, Springer, 2011 [8] J.-R Wen, J.-Y Nie, and H Zhang Query clustering using user logs ACM Transactions on Information Systems, Vol 20, No 1, pages 59 81, 2002 68 Ph l c K t qu kh o sát: nhóm hay khơng? bao nhiêu? - 69 Ph l c có quan KS_TTTK_001 KS_TTTK_002 KS_TTTK_003 KS_TTTK_004 KS_TTTK_005 KS_TTTK_006 KS_TTTK_007 KS_TTTK_008 KS_TTTK_009 KS_TTTK_010 KS_TTTK_011 KS_TTTK_012 KS_TTTK_013 KS_TTTK_014 KS_TTTK_015 KS_TTTK_016 KS_TTTK_017 KS_TTTK_018 KS_TTTK_019 KS_TTTK_020 KS_TTTK_021 KS_TTTK_022 KS_TTTK_023 KS_TTTK_024 KS_TTTK_025 KS_TTTK_026 KS_TTTK_027 KS_TTTK_028 KS_TTTK_029 KS_TTTK_030 KS_TTTK_031 KS_TTTK_032 KS_TTTK_033 KS_TTTK_034 KS_TTTK_035 27 23 32 34 34 34 38 25 25 34 38 33 34 34 25 34 34 28 44 17 34 31 33 32 32 32 32 25 27 34 34 27 23 23 35 x x x x x x x x x x x x x x x x x x x x x x x x x x x x x X X x x x x Bn bán Bn bán Kinh Doanh Kinh Doanh 70 Khóa 0.9 0.95 0.8 0.85 1 0.8 0.9 0.95 0.99 0.8 0.85 0.85 0.95 0.95 0.99 1 0.95 0.9 0.9 0.85 1 0.9 0.9 0.7 0.8 0.7 0.8 Khóa 0.8 0.8 0.75 0.7 0.75 0.8 0.85 0.75 0.8 0.9 0.9 0.75 0.75 0.75 0.8 0.8 0.85 0.8 0.85 0.8 0.85 0.85 0.85 0.8 0.8 0.9 0.9 0.9 0.6 0.7 0.85 0.6 0.7 Khóa 0.6 0.6 0.7 0.6 0.65 0.75 0.8 0.7 0.75 0.85 0.7 0.6 0.65 0.6 0.7 0.7 0.7 0.6 0.7 0.7 0.75 0.75 0.7 0.75 0.6 0.9 0.7 0.8 0.7 0.8 0.5 0.5 0.75 0.5 0.65 Khóa 0.5 0.4 0.65 0.5 0.55 0.7 0.7 0.65 0.7 0.8 0.5 0.45 0.55 0.5 0.65 0.6 0.65 0.5 0.6 0.6 0.65 0.65 0.55 0.7 0.4 0.9 0.6 0.7 0.7 0.8 0.4 0.4 0.6 0.4 0.6 Khóa 0.45 0.2 0.6 0.4 0.45 0.65 0.6 0.6 0.65 0.7 0.3 0.3 0.45 0.4 0.45 0.5 0.5 0.4 0.5 0.5 0.55 0.55 0.4 0.65 0.2 0.8 0.5 0.5 0.5 0.7 0.3 0.3 0.5 0.3 0.5 Ph l c KS_TTTK_036 KS_TTTK_037 KS_TTTK_038 KS_TTTK_039 KS_TTTK_040 KS_TTTK_041 KS_TTTK_042 KS_TTTK_043 KS_TTTK_044 KS_TTTK_045 KS_TTTK_046 KS_TTTK_047 KS_TTTK_048 KS_TTTK_049 KS_TTTK_050 KS_TTTK_051 KS_TTTK_052 KS_TTTK_053 KS_TTTK_054 KS_TTTK_055 KS_TTTK_056 KS_TTTK_057 KS_TTTK_058 KS_TTTK_059 KS_TTTK_060 KS_TTTK_061 KS_TTTK_062 KS_TTTK_063 KS_TTTK_064 27 31 28 25 22 23 32 29 31 28 28 34 29 31 30 30 31 31 31 27 28 31 26 40 36 26 25 24 30 Nhân viên kinh doanh Nhân viên IT Nhân viên IT Hành Hành Giáo viên Trung bình x x x x x x x x x x x x x x x x x x x x x x x x x x x x x 0.9 0.85 1 0.9 0.7 0.8 0.85 1 0.9 0.9 0.8 0.85 0.95 0.8 0.9 0.85 0.9 0.95 0.8 0.9 0.95 0.9 0.6 0.75 0.8 0.9 0.8 0.8 0.6 0.6 0.75 0.9 0.8 0.7 0.8 0.7 0.7 0.85 0.8 0.7 0.8 0.75 0.9 0.8 0.85 0.75 0.7 0.9 0.85 0.7 0.8 0.91 0.79 71 0.5 0.6 0.7 0.6 0.7 0.7 0.5 0.5 0.65 0.8 0.6 0.6 0.5 0.6 0.65 0.75 0.7 0.6 0.7 0.65 0.8 0.7 0.7 0.7 0.6 0.8 0.75 0.6 0.7 0.4 0.5 0.5 0.5 0.6 0.6 0.3 0.4 0.55 0.6 0.4 0.5 0.4 0.5 0.6 0.65 0.6 0.5 0.6 0.55 0.7 0.6 0.5 0.6 0.5 0.7 0.65 0.5 0.6 0.3 0.4 0.4 0.3 0.5 0.4 0.3 0.3 0.45 0.5 0.2 0.4 0.3 0.4 0.5 0.55 0.4 0.4 0.5 0.45 0.6 0.4 0.4 0.5 0.4 0.6 0.55 0.4 0.5 0.67 0.56 0.45 Ph l c tách n i dung fanpage [=] class:name [Avatar] = src [=] itemprop:name [PageName] = InnerText [Likes] = InnerText [=] id:timelineNavContent 72 Ph l c [=] class:fbTimelineNavigation

[=] id:fbTimelineNavTopRow
• [=] class:about tile [=] class:detail frame [=] class:fbProfileNonIconBylineFragment [Group] = InnerText [=] class:fbLongBlurb [Description] = InnerText

73 Ph l c

1. [=] role:article [Article] = InnerText [Article] = InnerText [Article] = InnerText

[UserComments] = InnerText 74 Ph l c tách n i dung trang cá nhân l y fanpage m i [=] id:favorites

• [FavoritePage] = href

[FavoritePage] = href 75 Ph l c D li u t - c t m t fanpage Link: URL fanpage - rút trích thơng tin c a fanpage c ch nh - Update Date: Ngày t i d li u - Avatar: URL c a hình nh - PageName: Tên fanpage - Like: Thơng tin v s - Group: Thông tin v th lo - Description: Mô t v - Article: N - UserComment: Thông tin thành viên tham gia bình lu n i di n ng like fanpage i t o ch nh i t o cung c p i fanpage c cung c i d ng code [Link] https://www.facebook.com/ManVsWild [/Link] [Template] facebook_contents_template3 - Copy3 [/Template] [Update date]:15-11-2013 05:43:08 PM [Avatar] https://fbcdn-profile-a.akamaihd.net/hprofile-akprn1/c13.13.167.167/s160x160/225889_10150230852219884_7327091_n.jpg [/Avatar] [PageName] Man vs Wild [/PageName] [Likes] [/Likes] [Group] 76 Ph l c [/Group] [Description] Welcome to the official Man vs Wild Facebook page! Bear Grylls is back on Discovery in the all-new series, "Escape From Hell." [/Description] [Article] Knowing how to boost a signal fire could make the difference between life and death in a survival situation Watch Bear show you how to it using a first aid kit in this sneak peek of Bear Grylls: Escape From Hell http://bit.ly/1j4NKF6 [/Article] [Article] Man vs Wild [/Article] [Article] Check it out! All of Bear's cold weather survival tips from Bear Grylls: Escape From Hell all in one place [/Article] [Article] Check it out! All of Bear's cold weather survival tips from last night's Bear Grylls: Escape From Hell episode all in one place: http://bit.ly/17n0j9G [/Article] [Article] Thích [/Article] [Article] Watch Bear demonstrate how to bandage a broken leg using a backpack, reenacting an incredible real-life survival story from tonight's special premiere of Bear Grylls: Escape From Hell http://bit.ly/180o2A0 [/Article] [Article] What would you ask Bear Grylls? Now's your chance! Bear is doing an AMA (Ask Me Anything) right now! http://redd.it/1qe15g [/Article] [Article] True or False: In a winter survival situation, you should eat ice and snow to avoid dehydration How are your cold weather survival skills? Take our Bear Grylls: Escape From Hell quiz and find out: http://bit.ly/HRMtoM [/Article] [Article] "Sometimes, you need to break your own rules and trust your instincts A good adventurer knows when to that and when not to." [/Article] [Article] 77 Ph l c "Sometimes, you need to break your own rules and trust your instincts A good adventurer knows when to that and when not to." Get more of Bear's Survival Myths & Rookie Mistakes here: http://bit.ly/1cLatbG [/Article] [Article] Has everyone seen the sneak peek of the new series Bear Grylls: Escape From Hell? Tell us what you think! http://bit.ly/1hQh7wZ [/Article] [Article] Hey Gryllers! Make sure you Like the official Bear Grylls: Escape From Hell page for the latest updates! The new series begins Wednesday Nov 13th at 10/9c with a special premiere Monday Nov 11th at 10/9c You can learn more about #EscapeFromHell here: http://bit.ly/1hQh7wZ [/Article] [Article] Bear Grylls is back on Discovery! Don't miss a special premiere of his new series "Escape From Hell" Monday Nov 11th at 10/9c! [/Article] [Article] If you're lost in the wild with a busted cell phone and no means to make a fire, Bear Grylls has an idea for you (But only as a last resort.): [Video] http://bit.ly/1eQ7LNq [/Article] [UserComments] aggregatedranges":[{"offset":45,"length":17,"count":24958}],"ranges":[{"offset":5,"lengt h":13,"entities":[{"url":"https:\/\/www.facebook.com\/garima.sehgal.969","id":100005 415431150}]},{"offset":20,"length":3,"entities":[{"url":"","shimhash":"MAQH9CqhR"," id":100002439434543}]},{"offset":25,"length":16,"entities":[{"url":"","shimhash":"PAQ FmCUN","id":100005419176120}]}]}},"sharecount":965,"ownerid":"41036834883","last seentime":null,"canremoveall":false,"seencount":0,"seenbyall":false,"hasunseencollapsed ":false,"mentionsdatasource":{" m":"m_0_1d"},"isranked":false,"isthreaded":false,"iso wnerpage":true,"isownerauthoritativepage":true,"isshare":false,"isqanda":false,"actorid":" 41036834883","grouporeventid":null,"allowphotoattachments":false,"allowstickerattach ments":false,"replysocialsentencemaxreplies":10,"showfeaturedreplies":false,"viewercans ubscribetopost":false,"hasviewersubscribed":false,"commentboxhoisted":false,"hasaddco mmentlink":false,"defaultcommentorderingmode":"chronological"}],"comments":[{"id": "10152054865289884_11642594","fbid":"10152059651189884","legacyid":"11642594 [/UserComments] - 78 ... j cách 31 | i= 1,2, : Cácnghiên c u liên quan SM = SM = hai i i dùng ui j j k g uj SM2 = SM*SM 3.3 Khai phá liên k t web 32 : Cácnghiên c u liên quan khóa - trang trang i trang pj trang pi trang. .. nh n ch c a t ng fanpage t i , h th ng có th g i ý i dùng nh ng fanpage phù h p nh t 25 : lýthuy t facebook chúng -idf fanpage ùng Sau k-means tf- 26 : Cácnghiên c u liên quan CÁC ng khai phá... ng fanpage phát i dùng vi c l a ch n fanpage i dùng g p m g i ý b i facebook Nh n th cs ng fanpage l c i dùng nghiên c u xây d ng h th ng g i ý facebook fanpage theo tiêu chí c a i dùng v ng