(p1) Hierarchical combinations of CC-100 and Hspell

(p1) Filter CC-100 to reduce legal and commercial content (or consider other corpuses)

(p2) Rerank palindrome/anadrome lists by (joint) counts

(p2) Extract bigrams and counts from CC100

(p3) Split part-of-speech lists by base forms, gender and plurality

(p3) Use the above splits to find palindromic bigrams of sentence form (e.g. noun adj, noun verb) and matching gender and plurality (and perhaps definiteness)

(p4) Extract my own Bible wordlist

(p4) Split שמות-פרטיים.xlsx by gender and Jewishness

(p5) Palindrome/anadrome lists using ALL possible Hebrew prefixes

(p5) Allow CC-100 words to include acronyms and abbreviations

This file was written using Markdown Editor, the only live editor I found that respects html dir=rtl tags, while retaining markdown semantics inside the div

This is version 1.4 of Hspell, the free Hebrew spellchecker and morphology engine.

You can get Hspell from: http://hspell.ivrix.org.il/

Hspell was written by Nadav Har'El and Dan Kenigsberg: nyh @ math.technion.ac.il danken @ cs.technion.ac.il

Hspell is free software, released under the GNU Affero General Public License (AGPL) version 3. Note that not only the programs in the distribution, but also the dictionary files and the generated word lists, are licensed under the AGPL. There is no warranty of any kind for the contents of this distribution. See the LICENSE file for more information and the exact license terms.

The rest of this README file explains Hspell's spelling standard (niqqud-less), a bit about the technology behind Hspell, how to use the "hspell" program (but see the manual page for more current information), and lists a few future directions. See the separate INSTALL file for instructions on how to install Hspell.

Hspell was designed to be 100% and strictly compliant with the official niqqud-less spelling rules ("Ha-ktiv Khasar Ha-niqqud", colloquially known as "Ktiv Male", or "plene spelling" in English), published by the Academy of the Hebrew Language. This is both an advantage and a disadvantage, depending on your viewpoint. It's an advantage because it encourages a correct and consistent spelling style throughout your writing. It is a disadvantage, because a few of the Academia's official spelling decisions are relatively unknown to the general public.

Users of Hspell (and all Hebrew writers, for that matter) are encouraged to read the Academia's official niqqud-less spelling rules (which are printed at the end of most modern Hebrew dictionaries), and to refer to Hebrew dictionaries which use the niqqud-less spelling (such as Millon Ha-hove or Rav Milim). We also provide in docs/niqqudless.odt a document (in Hebrew) which describes in detail Hspell's spelling standard, and why certain words are spelled the way they are.

The "hspell" program itself is mostly a simple (but efficient) program that checks input words against a long list of valid words. The real "brains" behind it are the word lists (lexicon) provided by the Hspell project.

In order for it to be completely free of other people's copyright restrictions, the Hspell project is a clean-room implementation, not based on other companies' word lists, on other companies' spell checkers, or on copying of printed dictionaries. The word list is also not based on automatic scanning of available Hebrew documents (such as online newspapers), because there is no way to guarantee that such a list will be correct, complete, or consistent with regard to spelling rules.

Instead, our idea was to write programs which know how to correctly inflect Hebrew nouns and conjugate Hebrew verbs. The inputs to these programs are lists of noun stems and of verb roots, plus hints needed for the correct inflection when these cannot be figured out automatically. These input files are obviously an important part of the Hspell project. The "word list generators" (written in Perl, and are also part of the Hspell project) then create the complete word-list for use by the spellchecking program, hspell. The generated lists are useful for much more than spellchecking, by the way - see more on that below ("the future").

Although we wrote all of Hspell's code ourselves, we are truly indebted to the old-style "open source" pioneers - people who wrote books about the knowledge they developed, instead of hiding it in proprietary software. For the correct noun inflections, Dr. Shaul Barkali's "The Complete Noun Book" has been a great help. Prof. Uzzi Ornan's booklet "Verb Conjugation in Flow Charts" has been instrumental in the implementation of verb conjugation, and Barkali's "The Complete Verb Book" was used too.

During our work we have extensively used a number of Hebrew dictionaries, including Even Shoshan, Millon Ha-hove and Rav-Milim, to ensure the correctness of certain words. Various Hebrew newspapers and books, both printed and online, were used for inspiration and for finding words we still do not recognize. We wish to thank Cilla Tuviana and Dr. Zvi Har'El for their assistance with some grammatical questions.

After unpacking the distribution and running "configure", "make" and "make install" (see the INSTALL file for more information), the hspell executable is installed (by default) in /usr/local/bin, and the dictionary files are in /usr/local/share/hspell.

The "hspell" program can be used on any sort of text file containing Hebrew and potentially non-Hebrew characters which it ignores. For example, it works well on Hebrew text files, TeX/LaTeX files, and HTML. Running

hspell filename

Will check the spelling in filename and will output the list of incorrect words (just like the old-fashioned UNIX "spell" program did). If run without a file parameter, hspell reads from its standard input.

In the current release, hspell expects ISO-8859-8-encoded files. If files using a different encoding (e.g., UTF8) are to be checked, they must be converted first to ISO-8859-8 (e.g., see iconv(1), recode(1)).

If the "-c" option is given, hspell will suggest corrections for misspelled words, whenever it can find such corrections. The correction mechanism in this release is especially good at finding corrections for incorrect niqqud-less spellings, with missing or extra 'immot-qri'a.

The "-l" (verbose) option will explain for each correct word why it was recognized, if Hspell was built with the "linginfo" optional feature enabled (a morphological analysis is shown, i.e., fully describe all possible ways to read the given word as an inflected word with optional prefixes).

Because hspell's output (naturally) is "logical-order", it is normally useful to pipe it to bidiv or rev before viewing. For example

hspell -c filename | bidiv | less

Another convenient alternative is to run hspell on a BiDi-enabled terminal.

Instead of using the hspell program described above, users can also use Hspell's lexicon through one of the popular multi-lingual spell-checkers, aspell and hunspell. See the INSTALL file for more information on building these dictionaries.

By now, Hspell is fairly mature, and its lexicon of over 24,000 base words is fairly comprehensive, similar in breadth to some printed dictionaries. Careful attention has also been given to its accuracy, and its conformance with the spelling rules of the Academy of the Hebrew Language.

Nevertheless, Hspell does not, and probably never will, cover all of modern Hebrew language. Also, undoubtedly, it may contain some errors as well. If you find such omissions or errors, please let us know.

Before reporting such omissions or errors, please try to verify that the word you are proposing is indeed correctly spelled: Please refer to modern dictionaries. Please also look at doc/niqqudless.odt - the word you are proposing might actually be a known mispelling which we discuss in that document.