{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

lecture7-vectorspace-handout-6-per

# G nd docs with merchant in the tle zone and matching

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: al Follower Sec. 7.1.6 Why use random sampling General variants   Fast   Leaders reﬂect data distribu*on   Have each follower afached to b1=3 (say) nearest leaders.   From query, ﬁnd b2=4 (say) nearest leaders and their followers.   Can recurse on leader/follower construc*on. 6 Introduc)on to Informa)on Retrieval Sec. 7.1.6 Introduc)on to Informa)on Retrieval Sec. 6.1 Exercises Parametric and zone indexes   To ﬁnd the nearest leader in step 1, how many cosine computa*ons do we do?   Thus far, a doc has been a sequence of terms   In fact documents have mul*ple parts, some with special seman*cs:   Why did we have √N in the ﬁrst place?   What is the eﬀect of the constants b1, b2 on the previous slide?   Devise an example where this is likely to fail – i.e., we miss one of the K nearest docs.   Likely under random sampling.             Author Title Date of publica*on Language Format etc.   These cons*tute the metadata about a document Introduc)on to Informa)on Retrieval Sec. 6.1 Introduc)on to Informa)on Retrieval Sec. 6.1 Fields Zone   We some*mes wish to search by these metadata   A zone is a region of the doc that can contain an arbitrary amount of text, e.g.,   E.g., ﬁnd docs authored by William Shakespeare in the year 1601, containing alas poor Yorick   Year = 1601 is an example of a ﬁeld   Also, author last name = shakespeare, etc.   Field or parametric index: pos*ngs for each ﬁeld value   Some*mes build range trees (e.g., for dates)   Field query typically treated as conjunc*on   Title   Abstract   References …   Build inverted indexes on zones as well to permit querying   E.g., ﬁnd docs with merchant in the *tle zone and matching the query gentle rain   (doc must be authored by shakespeare) Introduc)on to...
View Full Document

{[ snackBarMessage ]}