0 0 votes Consider the following statements 1. Decomposition of a relation into BCNF may not be lossless. 2. If R and S are two relations in BCNF the natural join of R and S is also in BCNF. which of the following is correct ? a) only 1 b) only 2 c) Both 1 and 2 d) None of these I have doubt in first option..please exlain.. Databases + – Siddharth Bhardawaj 2.1k views answer comment Share Follow Print See all 15 Comments 15 15 Comments reply Shaik Masthan commented Sep 22, 2018 reply Follow flag we can guarantee that at least one decomposition which gave loss-less in BCNF but we can't guarantee that at least one decomposition which gave Dependency preserving in BCNF. What is he answer for 2nd statement, ( i hope, it should be some more specific, i mean they didn't mention that those are having common attribute or not? etc... ) 0 0 replyShare Siddharth Bhardawaj commented Sep 22, 2018 reply Follow flag answer is option (a)... but in the first statement , we check the decomposition of BCNF , the condition must satisfy:- 1. lossless join 2. Dependency preservation and then it should satisfy BCNF... so by going this , how it is possible for a BCNF decomposition may or may not be lossless... 0 0 replyShare BASANT KUMAR commented Sep 22, 2018 reply Follow flag i think if there is one common attribute in it then it will remain in BCNF for ex R(A,B) and S(B,C) then its natural join will be in BCNF.but i am confused what will happen if there is no any common attribute ??it will become cross product of two relation but we can say that it is in BCNF??if yes how??if no how?? 0 0 replyShare Shaik Masthan commented Sep 22, 2018 reply Follow flag from where you get this questions, even at least one question is also not standard. ( have ambiguity ) 0 0 replyShare Siddharth Bhardawaj commented Sep 22, 2018 reply Follow flag Made easy workbook questions.. 0 0 replyShare Shaik Masthan commented Sep 22, 2018 reply Follow flag we check the decomposition of BCNF , the condition must satisfy:- 1. lossless join 2. Dependency preservation sometimes it may not satisfy Dependency preservation. Made easy workbook questions.. is it extra work book? then don't solve questions from it 0 0 replyShare BASANT KUMAR commented Sep 22, 2018 reply Follow flag option (a) may be wrong consider the situation R(A,B) and S(B,C) number of tuple in R>>S if we take natural join then it may be possible that some of the tuple of relation R which satisfy A->B get lost due to natural join.so it can't satisfy lossless .correct me if my approach is wrong. 0 0 replyShare Siddharth Bhardawaj commented Sep 22, 2018 reply Follow flag kkk...@shaik masthan 0 0 replyShare himgta commented Sep 23, 2018 reply Follow flag @Shaik Masthan what about 1st statement....is it true? 0 0 replyShare Shaik Masthan commented Sep 23, 2018 reply Follow flag @himgta it is also ambiguity question ( i mean question need to some more specific ) 0 0 replyShare himgta commented Sep 24, 2018 i moved by himgta Sep 24, 2018 reply Follow flag @Shaik Masthan I just want to know whether the decomposition in BCNF will always be lossless or it can be lossy sometimes 0 0 replyShare Prateek Raghuvanshi commented Sep 25, 2018 reply Follow flag Decomposition of a relation into BCNF always be lossless. 0 0 replyShare Shaik Masthan commented Sep 25, 2018 reply Follow flag @himgta if you want you can have a loss-less decomposition always, but note that every decomposition of BCNF doesn't need not to satisfy Lossless property. R(A,B,C,D) with FD set ={AB → D, C → AD} i want to decompose it in R1(A,B,D) R2(A,C,D), it is in BCNF and Dependency Preservation but lossy R1(A,B,D) R2(A,C,D), and R3(B,C) ===> BCNF, DP and Loss-less 0 0 replyShare Prateek Raghuvanshi commented Sep 25, 2018 reply Follow flag @ Shaik Masthan but at least one decomposition of that relation into BCNF always be lossless ,not any particular given decomposition. 0 0 replyShare Shaik Masthan commented Sep 26, 2018 reply Follow flag @Prateek Raghuvanshi yes brother, i am also conveyed that one only. 0 0 replyShare Please log in or register to add a comment.