Blog

Articles on automated cataloging, catalog formats and use cases.

How to generate MARC21 and UNIMARC records with AI

How Janium Collect generates MARC21 and UNIMARC records with AI, when each is used, and how it delivers that record as MarcEdit text, MARCXML, binary MARC or a flat file (FLAT), according to what the receiving catalog expects.

Janium 2027 is now available

Janium 2027 adds a relevance-ranked search engine, file display in the OPAC, background inventory, an email sign-in code and password expiry.

From the image and the catalog card to the record

How to combine an object's image and its catalog card to catalog: Collect takes what is visible in the reproduction and the documentary data from the card, and produces a record. What neither source supports is not presented as verified.

From any format to a catalog record

Which document formats AI can catalog with Janium Collect: PDF, images and plain text. From a PDF come the catalog record and structured markdown for full text, display and, if needed, RAG.

Integrating AI cataloging with your catalog (ILS)

How Collect records reach a catalog or ILS —load batch, SFTP, folder— and what that integration does not do: Collect does not keep a synchronized copy of every record. On hierarchical ISAD-G the catalog sends Collect the tree and the dates for bubble-up.

Cataloging audiovisual collections — newsreels, recordings and video

How Janium Collect helps describe an audiovisual collection at scale —newsreels, sound recordings, video records of events— from the transcription and the per-scene frames, producing a parent record and child segments in the format the institution uses.

Keeping a degraded batch out of the catalog: quality gates

How to keep a degraded batch out of the catalog: Collect checks cloud sources at start-up and stops the load if links to the original or a required record type were lost. Fail in the open; do not silently ship bad data.

Authority control: each name enters the catalog in one form

The same author spelled three ways is three people in the index. Collect checks each name and subject against the lists the institution already uses —its catalog, VIAF, ISNI, the Getty vocabularies— and records which list the form came from.

Cataloging a collection video with AI: from the recording to linked records

A newsreel or a recorded session does not come with the content written down. Collect proposes a record for the whole video and, if needed, a record per scene, linked to the first.

Access points, full text and RAG

A recent study measures how much context a model actually uses. What that means for RAG, and what an institution gets when Collect surfaces access points —including the corporate extensions— for Janium to search.