Our Solr Subscription Package (Package) is a latest stable version of Apache Solr uniquely packaged by RONDHUIT in order for enterprise users to utilize the software safely and easily.
As Solr support service, of course, is provided with up-to-date knowledge and experience that RONDHUIT consultants acquired through their real-world consulting experience, you can focus on developing your business logic.

Easy Setup

The package includes a stable version of Solr that supports Japanese and enables you to start using it at once by simply running the installation script. Also, you can use and study the latest search technology as you read through the accompanying Japanese manuals because sample environments and sample data for a wide variety of usage situations are included.

Finely Textured Japanese Processing Plug-ins

Japanese documents have unique “spelling inconsistency” and it could cause “missed search” unless it is handled properly . This “missed search” could lead to serious situation depending on the type of your business operation. For example, “missed search” during web site search on EC sites could lead to missed potential customers that cause lost profit in the future. RONDHUIT’s Japanese processing plug-ins absorb unique Japanese spelling inconsistency like the followings in order to prevent missed search from occurring.

One-byte/two-byte

ｱｲｳ１２３⇔アイウ123

Old/New Chinese characters

慶應大学⇔慶応大学

Iteration mark

時々⇔時時、部分々々⇔部分部分、いすゞ自動車⇔いすず自動車

Chinese numerals and Arabic numerals

四七／四十七／四拾七⇔47

Phonetic notation

かたかな⇔カタカナ⇔katakana、日本語⇔にほんご

Compound loan words

オープンソース・ソフトウェア⇔オープンソースソフトウェア

Geographical names/Conjugational characters

旭が丘⇔旭丘／卸売り⇔卸売／下請け⇔下請／垂れ幕⇔垂幕

Extracting and Getting the Most out of Your Business Buzzwords

When you build a system that searches documents such as Word and PDF that include many “natural sentences”, Solr features including facet/clustering/similar document search would become difficult to perform their functions because there are not enough metadata of documents being provided. You can use the term extraction function that is included in the package to extract “feature words” – i.e. “metadata” – and feed them back to the documents and get the most out information the documents originally have. This enables you to fully utilize the facet/clustering/similar document search functions and suggestion (see the following diagram).

Japanese-suported “Did you mean” Search

When search word errors occur as a result of keyboard typo and/or wrong kanji conversion, the system presumes similar search words and could display “Did you mean…” option to the user. A correct keywords would be displayed and the user could simply click the link to perform another search leading to work saving.

Support Service

We will provide customers, who use our subscription package to develop and/or manage a search system, with solutions including bug fixes and work-around suggestions that best fit their situations. The provided bug fixes are basically fed back to Apache Lucene/Solr community in order to bring them to commit status. This way, we can make sure to prevent a similar failure from occurring in future versions. This is the support service that only a company like RONDHUIT, which has Apache Lucene/Solr committers as its team members, could provide.