Infer

[Willgren]

Hi!

 

I am currently using the [infer] function. Brilliant! I'm just not entirely clear of what parameters are actually at play. I use the default settings, ([iNFER 6 BHS-W4]), and get a lot of interesting hints, but what does the 6 stand for?

 

Let me take one example, for Ps 113, the first two hits are from 1 Sam 2:8 and Isa 9:6, but these are very different.

1 Sam 2:8 contains the long (important) verbatim similarity מֵקִים מֵעָפָר דָּל מֵאַשְׁפֹּת יָרִים אֶבְיוֹן

but Isa 9:6 only resembles the psalm by מֵעַתָּה וְעַד־עוֹלָם

 

I suppose that the number 6 in the latter case should refer to the preposition מ, the particle עתה, the conjunction ו, the preposition עד and the noun עולם. But that only makes it five? Or am I missing something?

 

The function is great, but it would be even greater if I knew exactly how to describe what I have searched for and what I have found . Any thoughts?

[Mike Thigpen]

Willgren,

 

I'd recommend looking at the Accordance help file on the Infer command. It is quite informative. The 6 in the default refers to the number of words in the phrases under consideration. You can change this number in the Infer dialog box.

 

Mike

[Willgren]

Thank you for your answer, Mike. I actually read through the help file before posting, but it did not help me. Maybe I'm just on a too basic level. But as it says there, the number specifies the words to be searched in a phrase, but the next sentence claims that the number specifies the length of the phrase. These two statements, combined with the search result I presented above (the words highlighted in red counts to 5, not 6, and these 5 do not constitute a phrase, but form a part of a phrase) makes me a bit confused. How would you explain your own explanation that "the 6... refers to the number of words in the phrases under consideration"? How would you count the number of "words" found in Is 9:6?

[Mike Thigpen]

The way I understand this is that the number specifies the number of words that will be consider together as a phrase in the source text. So it considers the words in the source text in groups of 6 and then compares those phrases to the match text. The results may return a phrase with less than the full number of words in it as a match. In the advanced options you have more control over how far the matched phrases can deviate from the source phrases.

 

The help text doesn't specify what the default parameters are for how much of the source phrase can be omitted and a match still be found. Hopefully others will chime in and can help with the default specifics.

[Helen Brown]

The default is that one word can be dropped and one added in the found phrase. If you get the dialog when you enter INFER from the menus, you will see many other options that you can specify, and yes, they are detailed in the Help files.