# understand the queryNorm and the fieldNorm.

5 messages
Open this post in threaded view
|
Report Content as Inappropriate

## understand the queryNorm and the fieldNorm.

 Hi, I have a problem of understanding the queryNorm and fieldNorm. The following is an example. I try to follow what said in the Javadoc "Computes the normalization value for a query given the sum of the squared weights of each of the query terms". But the result is different. ID:0 C:/PDF2Text/SearchEngine/File/SIG/sigkdd/p374-zhang.pdf|initial rank: 0 0.31900567 = sum of:   0.03968133 = weight(contents:associ in 920), product of:     0.60161763 = queryWeight(contents:associ), product of:       1.326625 = idf(docFreq=830)       0.45349488 = queryNorm     0.065957725 = fieldWeight(contents:associ in 920), product of:       4.2426405 = tf(termFreq(contents:associ)=18)       1.326625 = idf(docFreq=830)       0.01171875 = fieldNorm(field=contents, doc=920)   0.27932435 = weight(contents:rule in 920), product of:     0.7987842 = queryWeight(contents:rule), product of:       1.7613963 = idf(docFreq=537)       0.45349488 = queryNorm     0.34968686 = fieldWeight(contents:rule in 920), product of:       16.941074 = tf(termFreq(contents:rule)=287)       1.7613963 = idf(docFreq=537)       0.01171875 = fieldNorm(field=contents, doc=920) regards jiang xing
Open this post in threaded view
|
Report Content as Inappropriate

## Re: understand the queryNorm and the fieldNorm.

 Hi Jason, I get the same thing for the queryNorm when I calculate it by hand: 1/((1.7613963**2 + 1.326625**2)**.5)  = 0.45349488111693986 -Yonik On 2/6/06, jason <[hidden email]> wrote: > Hi, > > I have a problem of understanding the queryNorm and fieldNorm. > > The following is an example. I try to follow what said in the Javadoc > "Computes the normalization value for a query given the sum of the squared > weights of each of the query terms". But the result is different. > > ID:0 C:/PDF2Text/SearchEngine/File/SIG/sigkdd/p374-zhang.pdf|initial rank: 0 > 0.31900567 = sum of: >   0.03968133 = weight(contents:associ in 920), product of: >     0.60161763 = queryWeight(contents:associ), product of: >       1.326625 = idf(docFreq=830) >       0.45349488 = queryNorm >     0.065957725 = fieldWeight(contents:associ in 920), product of: >       4.2426405 = tf(termFreq(contents:associ)=18) >       1.326625 = idf(docFreq=830) >       0.01171875 = fieldNorm(field=contents, doc=920) >   0.27932435 = weight(contents:rule in 920), product of: >     0.7987842 = queryWeight(contents:rule), product of: >       1.7613963 = idf(docFreq=537) >       0.45349488 = queryNorm >     0.34968686 = fieldWeight(contents:rule in 920), product of: >       16.941074 = tf(termFreq(contents:rule)=287) >       1.7613963 = idf(docFreq=537) >       0.01171875 = fieldNorm(field=contents, doc=920) > > regards > jiang xing > > --------------------------------------------------------------------- To unsubscribe, e-mail: [hidden email] For additional commands, e-mail: [hidden email]
Open this post in threaded view
|
Report Content as Inappropriate

## time of search for an index with the file .FDT much large

 Hi,   I have an index with 2,5 million documents. A document is formed in this way: - 15 fields index - 1 field stored but not indexed, whose value is one string of 500 byte. A search in average gives back the 3000 document. As 3000 id of documents is given back a lot fastly, the 3000 documents instead demands at least 20sec. for being given back!!!   I have tried with a string of 5 byte instead that of 500, and the 3000 documents are given back in only 1sec!   The question seems had to the size of the file .fdt   How I can make in order to reduce the time in the first case?                     Ing. Antonio Bruno   Software Analyst   http://xoomer.virgilio.it/lnb  cell: (+39) 3402347684   T&S S.r.l - Technologies And Solutions   email T&S: [hidden email]   email Yahoo: [hidden email]                 --------------------------------- Yahoo! Mail: gratis 1GB per i messaggi, antispam, antivirus, POP3
Open this post in threaded view
|
Report Content as Inappropriate

## Re: time of search for an index with the file .FDT much large

 20 seconds does seem like a long time to retrieve the stored fields of the 3000 documents.  However, you should also step back and determine if you really need to do that, or if there is another way to narrow the number of documents that need to be read from disk. -Yonik On 2/6/06, Antonio Bruno <[hidden email]> wrote: > Hi, >   I have an index with 2,5 million documents. > A document is formed in this way: > - 15 fields index > - 1 field stored but not indexed, whose value is one string of 500 byte. > A search in average gives back the 3000 document. As 3000 id of documents is given back a lot fastly, the 3000 documents instead demands at least 20sec. for being given back!!! >   I have tried with a string of 5 byte instead that of 500, and the 3000 documents are given back in only 1sec! >   The question seems had to the size of the file .fdt >   How I can make in order to reduce the time in the first case? > > > > > > >   Ing. Antonio Bruno >   Software Analyst >   http://xoomer.virgilio.it/lnb>   cell: (+39) 3402347684 >   T&S S.r.l - Technologies And Solutions >   email T&S: [hidden email] >   email Yahoo: [hidden email] > > > > > > > > > --------------------------------- > Yahoo! Mail: gratis 1GB per i messaggi, antispam, antivirus, POP3 > --------------------------------------------------------------------- To unsubscribe, e-mail: [hidden email] For additional commands, e-mail: [hidden email]