[{"data":1,"prerenderedAt":157},["ShallowReactive",2],{"authors":3,"research-corpus-explorer-en":41},{"id":4,"authors":5,"extension":37,"meta":38,"stem":39,"__hash__":40},"authors\u002Fauthors.yaml",[6,11,15,20,25,29,33],{"id":7,"name":8,"last_name":9,"link":10},"prudence-castelot","Prudence","Castelot","",{"id":12,"name":13,"last_name":14,"link":10},"clemence-farenc","Clémence","Farenc",{"id":16,"name":17,"last_name":18,"link":19},"adrien-chuttarsing","Adrien","Chuttarsing","https:\u002F\u002Fadrienchuttarsing.com\u002F",{"id":21,"name":22,"last_name":23,"link":24},"hugo-saunier","Hugo","Saunier","https:\u002F\u002Fwww.instagram.com\u002Fmotodroitversleciel\u002F",{"id":26,"name":27,"last_name":28,"link":10},"francois-bobinet","François","Bobinet",{"id":30,"name":31,"last_name":32,"link":10},"vanessa-helou","Vanessa","Helou",{"id":34,"name":35,"last_name":36,"link":10},"sabiya-mapaka","Sabiya","Mapaka","yaml",{},"authors","hi1M2xDA2hyLM42CGwrkvFmPxHwNHxSU-TH-bxMFEZI",{"id":42,"title":43,"acknowledgement":44,"authors":45,"bibliography":46,"body":47,"cover_credits":60,"cover_img_url":144,"cover_legend":145,"cover_url":99,"description":53,"extension":146,"meta":147,"navigation":148,"path":149,"release_date":150,"secondary_authors":151,"seo":152,"slug":153,"stem":154,"sub_title":155,"__hash__":156},"research_en\u002Fresearch\u002Fcorpus-explorer.md","Corpus Explorer","Special thanks to Leonardo Impett and Violaine Boutet de Monvel for trusting us.",[16,7],[],{"type":48,"value":49,"toc":141},"minimark",[50,54,57,63,68,71,75,79,82,86,90,93,96,100,104,107,111,115,118,122,130,133],[51,52,53],"p",{},"This year, we spent six months working with the Bibliotheca Hertziana in Rome to develop a digital tool for exploring archives and to imagine new interfaces supporting the work of the art historians at the Machine Visual Culture Lab.",[51,55,56],{},"Now that collections run to hundreds of thousands of documents, the eye can no longer go through them one by one. Researchers need tools that show the whole without losing the detail, that bring out series, patterns and divergences, while leaving them in charge of their own criteria. The Corpus Explorer was born from that need.",[58,59],"paper-video",{"credits":60,"legend":61,"src":62},"Filmatters","Zooming into the corpus","https:\u002F\u002Fapp.filmatters.org\u002Fdata\u002Fresearch\u002Fcorpus-explorer\u002Floading-zoom.mp4",[64,65,67],"h1",{"id":66},"a-space-for-exploring-archives","A space for exploring archives",[51,69,70],{},"The Corpus Explorer is a digital space for exploring and analysing archives, combining the distant viewing of a large body of data with the precision of each user's own ways of seeing.",[58,72],{"credits":60,"legend":73,"src":74},"Searching the corpus","https:\u002F\u002Fapp.filmatters.org\u002Fdata\u002Fresearch\u002Fcorpus-explorer\u002Fsearch.mp4",[64,76,78],{"id":77},"documents-by-the-hundreds-of-thousands","Documents by the hundreds of thousands",[51,80,81],{},"Designed for researchers, laboratories, archives, museums and documentalists, it can host all kinds of corpora: images, photographs, films and sounds, by the hundreds of thousands.",[58,83],{"credits":60,"legend":84,"src":85},"Loading a corpus","https:\u002F\u002Fapp.filmatters.org\u002Fdata\u002Fresearch\u002Fcorpus-explorer\u002Floading.mp4",[64,87,89],{"id":88},"tailored-modes-of-analysis","Tailored modes of analysis",[51,91,92],{},"Its modes of analysis are finely tuned to fit each user.",[51,94,95],{},"The AI models it relies on are interchangeable, which brings their specific biases into view.",[58,97],{"credits":60,"legend":98,"src":99},"Cross-referenced analytical filters","https:\u002F\u002Fapp.filmatters.org\u002Fdata\u002Fresearch\u002Fcorpus-explorer\u002Ffiltres.mp4",[64,101,103],{"id":102},"select-extract","Select, extract",[51,105,106],{},"In a few clicks, it also lets you select documents and extract usable data from them.",[58,108],{"credits":60,"legend":109,"src":110},"Selecting and downloading","https:\u002F\u002Fapp.filmatters.org\u002Fdata\u002Fresearch\u002Fcorpus-explorer\u002Fselect-download.mp4",[64,112,114],{"id":113},"the-digital-art-historians-desk","The digital art historian's desk",[51,116,117],{},"The Corpus Explorer is conceived as the new desk of the digital art historian: a virtual space for bringing archives to light, and a place to reflect on how artificial intelligence models are adopted in the analysis, indexing and use of digital archives.",[64,119,121],{"id":120},"the-corpus-and-the-lab","The corpus and the lab",[51,123,124,125,129],{},"The dataset shown here is made of Italian newspapers (1901 to 1916), drawn from the research of Marta Pizzagalli (",[126,127,128],"em",{},"PhD student, Università della Svizzera italiana",").",[51,131,132],{},"This interface was developed within the Machine Visual Culture Lab (dir. Leonardo Impett, Max Planck Institute Bibliotheca Hertziana, Rome) for its researchers (Cambridge, UZH, etc.).",[51,134,135,140],{},[136,137,139],"a",{"href":138},"mailto:filmatters.research@gmail.com","Get in touch"," with any question: we are open to collaboration!",{"title":10,"searchDepth":142,"depth":142,"links":143},2,[],"https:\u002F\u002Fapp.filmatters.org\u002Fdata\u002Fresearch\u002Fcorpus-explorer\u002Fog-corpus-explorer.jpg","Italian newspapers (1901-1916), corpus by Marta Pizzagalli","md",{},true,"\u002Fresearch\u002Fcorpus-explorer","2026-10-25",[],{"title":43,"description":53},"corpus-explorer","research\u002Fcorpus-explorer","The digital art historian's desk for exploring and analysing corpora.","gTevKy2GEVSPf1_RwIu_sxaukAbbRff6Rk1C6c4Q_4w",1792930418836]