Indonesian Affix Counter

Count and classify ber-/me-/di-/ke- prefix usage in Indonesian text

Scans Indonesian text and tallies the major productive prefixes (me-, ber-, di-, ter-, ke-, pe-) including their allomorphs, so you can analyse affix usage at a glance. Free and instant. It runs free in your browser on Gera Tools, with nothing uploaded.

Last updated Source: Gera Tools

Which prefixes are counted?

The six major productive prefixes: me- (meN-), ber-, di-, ter-, ke-, and pe- (peN-). Each is matched through its surface forms, so meng-, mem-, and meny- all count toward me-.

The Indonesian Affix Counter scans Bahasa Indonesia text and reports how often each major prefix (imbuhan) appears. Indonesian is a strongly affixing language, and prefixes like me-, ber-, and di- carry grammatical meaning such as voice and verb formation. This tool gives teachers, linguists, and learners a quick statistical view of affix usage.

How it works

The counter checks each word against the surface forms of six productive prefixes:

  • me- (the meN- prefix): me-, mem-, men-, meng-, meny-, menge-
  • ber-: ber-, bel-, be-
  • di-: di-
  • ter-: ter-, te-
  • ke-: ke-
  • pe- (the peN- prefix): pe-, pem-, pen-, peng-, peny-, penge-

A prefix is only counted when a stem of at least three letters remains after it, which prevents the standalone prepositions di (at) and ke (to) from being miscounted as prefixes. Each word is attributed to a single leading prefix, so the per-prefix totals sum to the affixed-word count.

Example

In membaca the mem- form of meN- attaches to the root baca (read), so it counts toward me-. In diberikan the di- prefix counts toward di-, and bertanya counts toward ber-.

What each prefix signals grammatically

Understanding what you are measuring helps interpret the counts:

meN- (active verb) is the marker of active-voice transitive and intransitive verbs. A high me- count signals an active, actor-focused text — common in narrative prose, journalism, and dynamic writing where subjects perform actions.

di- (passive verb) marks passive voice in Indonesian — the grammatical object is promoted to subject and the actor is omitted or stated with oleh. A high di- count is characteristic of formal, bureaucratic, academic, or legal text, where passive constructions are strongly preferred. Comparing me- and di- counts gives a quick read of a text’s register.

ber- (stative or intransitive) forms verbs of state, condition, or activity, as in berbicara (to speak) and berlari (to run). It also appears in adjective-like predicates, so a high ber- count suggests descriptive or conversational text.

ter- (accidental, superlative, or potential) marks unintentional or involuntary actions (terjatuh, fell accidentally), superlatives (terbesar, biggest), and potential passives (terbaca, readable/able to be read). Its appearance is often a signal of evaluative or journalistic text.

ke- (mostly nominalisations) in modern Bahasa Indonesia forms abstract nouns from adjectives (kebersihan, cleanliness; keadilan, justice). A high ke- count points to abstract or policy-oriented writing.

peN- (agent nominalisations) forms nouns denoting the doer of an action (pembaca, reader; pengajar, teacher). Its frequency rises in academic texts describing roles and processes.

Practical uses

  • Language teaching: analysing a student essay’s me-/di- ratio reveals whether they are writing in the expected register
  • Text classification: news articles and government documents score very differently on di- density versus personal narratives
  • Readability research: texts with high total affix density tend to be denser and more demanding; lower density often correlates with simpler register
  • Content consistency: checking that translated or generated Indonesian text has a prefix distribution matching the target genre

Notes

  • Assimilation rules (meN- becoming meng- before vowels, mem- before b/p) are handled by listing all allomorphs.
  • Suffixes such as -kan and -an are not counted here; this tool focuses on prefixes.
  • Reduplication (e.g. berlari-lari) is handled as a single word matching its leading prefix.