The information collected in this system, such as search history and questionnaire results may be disclosed to the public for the improvement of the system and for certain academic purposes.
The intellectual property rights on all items of this site belong to the producers.
This regulation went into effect in April 1, 2016.
When you have finished reading all the articles in red letters on the left column, you can use this search system.
The data of this search engine was calculated based on the “BCCWJ” (Balanced Corpus of Contemporary Written Japanese, issued by the National Institute for Japanese Language and Linguistics)* and as a result, the most frequently used verbs and/or modifiers with the selected noun can be seen in this list. Therefore you can choose the most suitable collocation found within the native written language.
As some data in this corpus may have been modified for educational purposes, the producers of this website accept responsibility for the usage of data in this system. For details on data of this system, please visit the “Data of this System” page.
＊For details on the “BCCWJ”, please visit the following page by National Institute for Japanese Language and Linguistics: BCCWJ.
For reference purposes, this system provides information on nouns, verbs, and modifiers and about their corresponding former JLPT levels, along with their SVI for Japanese Learners proposed by Tokihiro (2016).
For the numbers such as “1-10” in this system, the first number represents the former JLPT level, while the second number represents its place on the SVI for Japanese Learners. Therefore, the numbers in this example shows that the word in question belongs to the level 1 of the old criteria of JLPT and level 10 of the SVI for Japanese Learners.
“SVI for Japanese Learners” is a term proposed by Dr. Yasuyo Tokuhiro (2006), which categorizes the vocabulary into ten levels according to the frequency and the word familiarity. The higher the number is, the more importance it has for Japanese learners because they are more frequently used among Japanese speakers. Note that the “Study Value Index for Japanese Learners” is designed for words of Chinese origin; words in Katakana are not indexed.
＊For details on the relationship between the former JLPT and the present one, please visit the following page: comparison (English)
＊For details on “Study Value Index for Japanese Learners”, please visit the following page: Study Value Index.
Many of the noun-verb collocations are translated. Each translation contains examples of the various possible meanings of the collocation.
All the example sentences of noun-verb collocation are created by native Japanese speakers and most of these have translations.
For the modifier-noun collocation, some of the most frequent words that follow are shown in the corpus instead of example sentences.
Additionally, you can discover the context in which the collocation is used more frequently, for instance: novels, websites, textbooks and white papers.
This search system is designed for Japanese learners to find the appropriate collocation in Japanese. A collocation is a sequence of words that appear more frequently than by simple chance. On this site, by typing or choosing a noun in the search engine, you can find the modifiers and verbs which occur more commonly with that noun.
This system is useful for when you need to know which modifier is more appropriate. For example, when modifying a noun, whether it is better to use 大きな (big), or 強い (strong) with the word 影響 (influence). Or, you can find out whether or not it is possible to use the verb あげる* with 影響 to mean “have an influence over [someone/something]”. *(In this case, 影響を与える is correct.)
This system enables you to use more suitable collocations, and helps your Japanese sound more natural.
・When is it Suitable to Use this System?
This system is designed for formal writing; for example, when writing a paper, report, and so on.
＊How to calculate the relationship (the frequency of appearance and the dice coefficient) between goiso entries in BCCWJ
In this system, “goiso” was used by the BCCWJ to calculate the frequency and the dice coefficient of a collocation. In the BCCWJ, the two verbs, “noru (乗る: to ride)” and “noru (載る: to appear in a newspaper, magazine etc.)”, share the same pronunciation and are grouped under the same goiso, represented by the kanji 乗る. Therefore, in this system, the frequency of the verb “noru (乗る)” is calculated using the goiso “乗る,” which is actually made up of the total amount of times that the verbs 乗る and 載る occur. The dice coefficient is also calculated based on the goiso’s frequency.
＊How to Select the Correct Kanji and its Appropriate Meaning － the Relationship between goiso in BCCWJ and the Kanji Used in this System －
As mentioned in the above example, the calculated frequency of the verbs grouped under the goiso “noru” includes the frequency of both “noru” (乗る, “to ride”, which co-occurs with “bus”) and the other “noru” (載る, “to appear in”, which is often found with the word “newspaper”). However, when the kanji of the verbs listed with their collocations are shown on the screen, the mistaken kanji 乗る has been adjusted to 載る where appropriate.
＊About Co-Occurrence and the Form of Verbs
In this system, when searching for collocations, the pattern retrieval method has been used for regular expressions; for example, to find the collocation of “noun + verb”, the pattern “noun + particle + (adverb +) verb” in “goiso” of the corpus was searched and counted. Therefore, the total amount of verbs includes the number of all the verb conjugations, along with their particles. For intransitive verbs “ga” is used, while for transitive verbs “o” is used.
The following sites in the “Nagareboshi” series are to assist in the learning of collocations, which are frequently used nouns and verbs. Please feel free to make use of them. These sites are free of charge and are suitable for mobile devices such as smartphones and tablets.
This work was supported by JSPS KAKENHI Grant Number 25370591.
We would like to thank for developing this mobile system.
We would also like to express our gratitude to Ninjal and Dr. Yasuyo Tokuhiro, for letting us use their data.