Use of Luke

ZZ Network

Open the index file with Luke for the first time. Overview displays the number of fields in the index. (the largest field is used, right ?), Is there the total number of documents and the term, that is, the total number of index words? Sometimes a keyword corresponds to multiple documents, so the number of documents is more than the number of keywords.

The version and format of the index are also displayed. Similar to the format of Java class, put the corresponding Lucene version information in the file header...

How is the rank in the lower right calculated? Actually, I didn't do anything....

In the document view,View the specific document items according to the doc number. This function is good.

You can use the document label to view each term. Click Show all to go to the search page, displaying all matching records using the term query.. The term here is equivalent to the query keyword...

The Luke query result is consistent with the direct code query result...

This feature is awesome. Through Luke, you can directly study the content in the index file .....

In the query tag, select the word divider .... Click "query... Here we select the CJK analyzer for East Asian text analyzer...

Using the same data source, we found the same thing now...

On the Search tab, click explain structure,You can see the word splitting result of the query string.....

(In the query input box) after updating the query string, you must click Update to update the following characters...

Each document corresponding to the term,This is equivalent to traversing the list.? Traversal from start to end? You can also show all. This will jump to the search tag ..You can also Click Show position to display the location information of the document..Here, term frequency refers to the number of times a term appears in the document..Only show all triggers the search and jumps to the search tag..

The following operation sequence is interesting.First click Next term to go to the next termClick "first doc" to display the first document of the term.

Documents corresponding to the same term are also sorted.? It is arranged according to the numbers after the document, at least from the above.

The tool menu bar has a Check Index option:

The following is my check operation on the current operation index...

Segments file = segments_2 numsegments = 1 version = format_has_prox [Lucene 2.4]
1 of 1: Name = _ 0 doccount = 153090.
Compound = true
Hasprox = true
Numfiles = 1
Size (MB) = 17.396
No Deletions
Test: Open reader... OK
Test: fields, norms... OK [2 fields]
Test: Terms, freq, prox... OK [88735 terms; 559179 terms/docs pairs; 563003 tokens]
Test: stored fields ...... OK [306180 total field count; AVG 2 fields per Doc]
Test: Term vectors... OK [0 total vector count; AVG 0 term/freq vector fields per Doc]

No problems were detected with this index.

OverviewShow top terms, Displays the term that corresponds to the largest number of documents...

You can use Luke to view the document scoring mechanism...

Numdocs indicates the total number of documents and docfreq indicates the total number of documents matching the term.IDF is defined above these two quantities.....

How are querynorm and fieldnorm calculated??

When you enable the index, you can choose to load the memory, which is much faster...