1. Presented By : Aamir Mushtaq Jesal Mistry Kapil Tekwani Neville Shah Visual Representation of Knowledge Articles as Dynamic Interactive Connected Graph Nodes Internal Guide: Prof. Mrs. Kalyani Waghmare External Guides: Mr. Prajwalit Bhopale Mr. Kiran Kulkarni Sponsored Organization: Infinitely Beta
2.
3.
4. Problem Definition To implement an Easy and Interactive E- Learning Tool for Knowledge Articles. It will be implemented as a browser plugin which will represent a graphical view of the document in the form of graphical nodes with main node focusing on keyword for which we want to gain information and neighboring nodes representing keywords that are most prominently related to the searched keyword/keyword about which information is to be obtained. In addition to that, we have semantic links between the nodes where the edges represent the relation.
5.
6.
7.
8. Algorithm Used (cont’d) 7. Depending on current depth, pre-decided window size to select top keyword/links for next level. Example: 20 for 0 th level, 10 for 1 st level, 5 for 2 nd level.(tuning required) 8. For efficient searching of accurate data we will be working across the depth i.e. at next levels if the keyword (present in previous level doc) is occurring many times (say 100), it will add weight to the corresponding keyword in the previous table. 9 Output will be graphical representation of keywords. If node (keyword) is a link, it will be connected to another node (keyword) of next level else stop at that level.
11. Mathematical Model Let S be the system. S = {U inp , U, D, Q, W t , K w , T Kw,S , T Kw,Wg , T U,Kw , T U,Kw,Wg } U inp = URL identifier (input to the system) D = database of the WWW, containing webpages as documents d i . D = {d 1 , d 2 , d 3 ,..., d n } where d i is a WWW document (webpage). Q = set of all possible queries. Q = {q 1 , q 2 , q 3, ..., q n } where q i is any given query to be fired on the database. W t = set of words of a particular document. W t = {w 1 , w2,..., wn} where w i ϵ d i, for 1<= i <= n K w = set of keywords ⊆ W t, obtained after F el K w = {k 1 , k 2 ,…, k m } where k i ⊆ W t , for 1<= i <= m U = extracted URLs from document d i U = {u 1 , u 2 ,..., u n } where u i ϵ d i
12. Mathematical Model T Kw,S = table of keywords and sectional counts, obtained after F cnt T Kw,S = {<k 1 , sA 1 , sB 1 , sC 1 >, <k 2 , sA 2 , sB 2 , sC 2 >, …, <k m , sA 3 , sB 3 , sC 3 >} T Kw,Wg = table of keywords and associated weights, obtained after F w T Kw,Wg = {<k 1 ,wg 1 >, <k 2 , wg 2 >, … ,<k m , wg m >} T U,Kw = table of urls in U mapped with the keywords and weights table T Kw,Wg obtained after F map T U,Kw = {<u n , k m , wg m >} U t is a mapping of keywords and their respective <U>
13. Mathematical Model Functions: F el (W T {<w 1 , w 2 , ... , w n >}) = K W F el eliminates all natural language elements from the <W T > part and resultant set of words are the keywords that are identified in the <K W > list / set. F cnt ( K w {<k 1 , k 2 , ... , k n >}) = T Kw,S F cnt returns an array of tuples of keywords and their respective sectional counts {<k m , s1, s2, s3>} which would be used in the calculation of weights of keywords. And provide the T Kw,S as input of F w . F w ( T Kw,S {<k m , sA m , sB m , sC m >}) = T Kw,Wg F w takes the T Kw,S obtained by the function F cnt as input and calculates the weight associated with each keyword and returns array of tuples of keywords and weights. {<k m , wg m >} F map ( U{<u 1 , u 2 , … u n >} ,T Kw,Wg {<k 1 ,wg 1 >, <k 2 ,wg 2 > ,…,<k m , wg m >}) = T U,Kw,Wg F map takes the U< u 1 , u 2 ...u n > and T Kw,Wg <k m , wg m > as input and it maps the keywords with the respective Urls in the d i and returns an array of urls with their mapped keywords and Urls. F win (lvl) = {<5> v <10> v <20>} F win is a window function that returns the size of the window that is dependent on the depth/ level that we are in.