The document discusses the challenges of indexing the Albanian language for search functionality. It does not have standardized lists of stop words or stemming rules like other languages. The author proposes mining texts to gather data on common words and suffixes to generate these resources for Albanian. An indexing process is outlined that would transliterate text, remove stop words and strip suffixes to standardize words before storing for search. Code for an initial algorithm is provided on GitHub.

1. Indexing the Albanian Language
by Andri Xhitoni
may 2015

2. Indexing the Albanian Language
The problem: Have search functionality
work on a website that's in Albanian.

3. Indexing the Albanian Language
The problem: Have search functionality
work on a website that's in Albanian.

4. Indexing the Albanian Language
The problem: Have search functionality
work on a website that's in Albanian.

5. Intricacies of search

6. Intricacies of search
Many think of search as
a straight-forward process

7. Intricacies of search
Many think of search as
a straight-forward process
“in go search terms, out come results”

8. it’s not that simple...
“in go search terms, out come results”

9. it’s not that simple...

10. it’s not that simple...
Words take on many forms.

11. it’s not that simple...
Words take on many forms.
Words may have different meanings
based on context

12. it’s not that simple...
Words take on many forms.
Words may have different meanings
based on context
Some words have no real semantic value
and must be ignored (stop words)

13. How do the big guys do it?

14. How do the big guys do it?
No searching through raw content

15. How do the big guys do it?
No searching through raw content
Search through optimized versions
of the raw content (indexing)

16. Basic indexing process
Alice was beginning to get very tired of sitting by her sister on
the bank, and of having nothing to do: once or twice she had
peeped into the book her sister was reading, but it had no
pictures or conversations in it, `and what is the use of a book,'
thought Alice `without pictures or conversation?'

17. Basic indexing process
Normalize the characters (transliteration)
and remove punctuation
alice was beginning to get very tired of sitting by her sister on
the bank, and of having nothing to do: once or twice she had
peeped into the book her sister was reading, but it had no
pictures or conversations in it, `and what is the use of a book,'
thought alice `without pictures or conversation?'

18. Basic indexing process
Remove stop words
alice was beginning to get very tired of sitting by her sister on
the bank, and of having nothing to do: once or twice she had
peeped into the book her sister was reading, but it had no
pictures or conversations in it, `and what is the use of a book,'
thought alice `without pictures or conversation?'

19. Basic indexing process
Transform each remaining word to its "basic version"
(stemming)
alice was beginning to get very tired of sitting by her sister on
the bank, and of having nothing to do: once or twice she had
peeped into the book her sister was reading, but it had no
pictures or conversations in it, `and what is the use of a book,'
think alice `without pictures or conversation?'

20. Basic indexing process
Store the indexed content alongside the original
alice was beginning to get very tired of sitting by her sister on
the bank, and of having nothing to do: once or twice she had
peeped into the book her sister was reading, but it had no
pictures or conversations in it, `and what is the use of a book,'
think alice `without pictures or conversation?'

21. Performing the search

22. Performing the search
the book Alice’s sister was reading

23. Performing the search
the book alice’s sister was reading
Perform the same indexing on the search terms

24. Performing the search
Search for the indexed search terms
in the indexed content
alice was beginning to get very tired of sitting by her sister on
the bank, and of having nothing to do: once or twice she had
peeped into the book her sister was reading, but it had no
pictures or conversations in it, `and what is the use of a book,'
think alice `without pictures or conversation?'
the book alice’s sister was reading

25. Performing the search
Rank results according to number of occurrences,
closeness of terms, position in the indexed text
alice was beginning to get very tired of sitting by her sister on
the bank, and of having nothing to do: once or twice she had
peeped into the book her sister was reading, but it had no
pictures or conversations in it, `and what is the use of a book,'
think alice `without pictures or conversation?'
the book alice’s sister was reading
2 21 1

26. Add the Albanian language
on top of the problem

27. Add the Albanian language
on top of the problem
No known "stop words" list

28. Add the Albanian language
on top of the problem
No known "stop words" list
Non-trivial stemming process

29. Add the Albanian language
on top of the problem
No known "stop words" list
Non-trivial stemming process
High irregularity in word formation

30. Add the Albanian language
on top of the problem
No known "stop words" list
Non-trivial stemming process
High irregularity in word formation
Vast number of forms for each single word

31. Just a taste of the complexity
Nouns 6 cases
x 2 numbers (singular, plural)
x 2 definitenes (definite, indefinite)
~24 word forms
Verbs 3 unique word-forming modes (of 6)
x 4 unique word-forming tenses (of 8)
x 2 voices (active, passive)
x 6 conjugative forms
~70 word forms

32. Looking for solutions

33. Looking for solutions
Ideally:

34. Looking for solutions
Ideally:
A list of stop words

35. Looking for solutions
Ideally:
A list of stop words
A (huge) list of all possible word forms
for all words in Albanian,
linked to their stem form.

36. Looking for solutions
Sources:

37. Looking for solutions
Sources:
The Dictionary
highly comprehensive
only base word forms

38. Looking for solutions
Sources:
The Dictionary
highly comprehensive
only base word forms
The Internet
not too comprehensive
many word forms
potential errors

39. Looking for solutions
Sources:
The Dictionary
highly comprehensive
only base word forms
The Internet
not too comprehensive
many word forms
potential errors
Hybrid source
a probability-based model
picking (hopefully) the best
from both sources

40. Data mining: Stop words

41. Data mining: Stop words
Get as many texts in Albanian as possible
(the more diverse, the better)

42. Data mining: Stop words
Get as many texts in Albanian as possible
(the more diverse, the better)
Transliterate the texts

43. Data mining: Stop words
Get as many texts in Albanian as possible
(the more diverse, the better)
Transliterate the texts
Keep a running count of the occurrence for each word

44. Data mining: Stop words
Get as many texts in Albanian as possible
(the more diverse, the better)
Transliterate the texts
Keep a running count of the occurrence for each word
Sort the list by occurrence count (highest first).

45. Data mining: Stop words
Get as many texts in Albanian as possible
(the more diverse, the better)
Transliterate the texts
Keep a running count of the occurrence for each word
Sort the list by occurrence count (highest first).
Stop words will float to the top.

46. Data mining: Stop words
Get as many texts in Albanian as possible
(the more diverse, the better)
Transliterate the texts
Keep a running count of the occurrence for each word
Sort the list by occurrence count (highest first).
Stop words will float to the top.
Manually white-list obvious false positives

47. Data mining: Stemming

48. Data mining: Stemming
Invert each word from the collected list

49. Data mining: Stemming
Invert each word from the collected list
Sort the list alphabetically
(effectively sorting by suffixes)

50. Data mining: Stemming
Invert each word from the collected list
Sort the list alphabetically
(effectively sorting by suffixes)
Find highest occurring suffixes of 2, 3 and 4 letters

51. Data mining: Stemming
Invert each word from the collected list
Sort the list alphabetically
(effectively sorting by suffixes)
Find highest occurring suffixes of 2, 3 and 4 letters
Manually look for false positives
and put them in a white list

52. The (basic) indexing algorithm

53. The (basic) indexing algorithm
Transliterate the input text

54. The (basic) indexing algorithm
Transliterate the input text
Find and remove all stop words

55. The (basic) indexing algorithm
Transliterate the input text
Find and remove all stop words
Go through each word and remove
the found suffixes (largest to smallest)

56. The (basic) indexing algorithm
https://github.com/andrixh/index-albanian
Transliterate the input text
Find and remove all stop words
Go through each word and remove
the found suffixes (largest to smallest)

57. Indexing the Albanian Language
by Andri Xhitoni
Thank you!
https://github.com/andrixh/index-albanian