README.md

In fauxneticien/anamR: Helper Functions for KDB

anamR

An R package of helper functions to read/write/process data from the Kaytetye database (aname is Kaytetye for 'helpless'). The lexicon is stored in a backslash-coded .txt file as shown below (first 10 lines):

 1 \lx ahe |1
 2     \lx_id 1
 3     \sk ah
 4     \hm 1
 5     \audio ahe
 6     \ta Stirling 450
 7     \sl MT JG
 8     \ps n.
 9     \sn 1
10         \de fight, quarrel, dispute, squabble, trouble
...
...

The function read.KDB() processes this text file into a table:

| lineno| indent|tag |content | lx_id| |------:|------:|:-----|:------------------------------------------|-----:| | 1| 0|lx |ahe |1 | 1| | 2| 4|lx_id |1 | 1| | 3| 4|sk |ah | 1| | 4| 4|hm |1 | 1| | 5| 4|audio |ahe | 1| | 6| 4|ta |Stirling 450 | 1| | 7| 4|sl |MT JG | 1| | 8| 4|ps |n. | 1| | 9| 4|sn |1 | 1| | 10| 8|de |fight, quarrel, dispute, squabble, trouble | 1| | ... | ... | ... | ... | ... |

Example

Get part of speech (ps tag) for each headword (lx, or lx_id)

library(anamR)
library(tidyr)

ps_df <- read.KDB("path/to/KDB.txt") %>%
            filter(tag == "ps")      %>%
            spread(tag, content)     %>%
            distinct(lx_id,
                     .keep_all = TRUE)

First 10 headwords and their parts of speech:

| lineno| indent| lx_id|ps | |------:|------:|-----:|:----| | 8| 4| 1|n | | 96| 4| 2|n | | 141| 4| 3|n | | 221| 4| 4|n | | 253| 4| 5|n | | 274| 4| 6|n | | 310| 4| 7|n | | 357| 4| 8|kin | | 382| 4| 9|kin | | 446| 4| 10|vi | | ... | ... | ... | ... |

Plot distribution of parts of speech in dictionary

library(ggplot2)

ps_df %>%
  group_by(ps) %>%
  tally()      %>%
  ggplot(aes(x = ps, y = n)) +
  geom_bar(stat = "identity")

Looks like there's a whole lot of nouns!

fauxneticien/anamR documentation built on May 16, 2019, 10:12 a.m.