Book part: Always Be Filtering
| dc.contributor.editor | Adelmann, Ralf | |
| dc.contributor.editor | Matzner, Tobias | |
| dc.creator | Apprich, Clemens | |
| dc.date.accessioned | 2024-07-24T11:08:34Z | |
| dc.date.available | 2024-07-24T11:08:34Z | |
| dc.date.issued | 2024 | |
| dc.description.abstract | Filters lie at the very heart of today’s machine learning systems. And because ma-chines process human – that is biased – data, these systems exacerbate existing inequalities in society and technology. In particular the homophilic principle of data analysis can be seen as a longstanding rule of segregation, buttressing most filter algorithms (e.g. recommendation systems). With algorithmically-enhanced sys-tems of pattern recognition, the question of how to publicly share information about these algorithms and investigate them for bias has become crucial. Instead of the still persistent assumption that algorithms are somehow hidden in a ‘black box’, inaccessible to human reasoning, the following contribution seeks to develop new forms of critique vis-à-vis filtering algorithms. Hence, a critical reflection of filtering mechanisms, which ultimately decide for us what to include and what to exclude, needs new approaches. It is therefore important to develop a hands-on approach, by confronting the theorisation of filters with actual machine learning techniques (e.g. by studying online machine learning courses). This will help to un-pack some of the mechanisms of algorithmic decision-making processes and build the basis for a genuine engagement with their real-world implications. | en |
| dc.identifier.doi | http://dx.doi.org/10.25969/mediarep/22939 | |
| dc.identifier.uri | https://mediarep.org/handle/doc/24440 | |
| dc.language | eng | |
| dc.publisher | Universität Paderborn | |
| dc.publisher.place | Paderborn | |
| dc.rights.uri | https://creativecommons.org/licenses/by/4.0/deed.de | |
| dc.subject | Algorithmic Filters | en |
| dc.subject | Machine Learning | en |
| dc.subject | Recommendation Systems | en |
| dc.subject | Critical Data Studies | en |
| dc.subject | Pattern Discrimination | en |
| dc.subject.ddc | ddc:300 | |
| dc.title | Always Be Filtering | en |
| dc.type | bookPart | |
| dc.type.status | publishedVersion | |
| dspace.entity.type | BookPart | |
| local.academicbookseries | Medienwissenschaftliches Symposium der DFG | |
| local.coverpage | 2024-07-25T02:38:54 | |
| local.source.booktitle | Filter | |
| local.source.epage | 10 | |
| local.source.spage | 1 | |
| local.source.volume | 4 | |
| relation.isAuthorOfBookPart | 5d4a78bd-1afd-4f12-a274-2ec9dfe99828 | |
| relation.isAuthorOfBookPart.latestForDiscovery | 5d4a78bd-1afd-4f12-a274-2ec9dfe99828 | |
| relation.isMiscOfBookPart | c3e086a7-382f-4afe-8a15-f6bda5fba6d5 | |
| relation.isMiscOfBookPart.latestForDiscovery | c3e086a7-382f-4afe-8a15-f6bda5fba6d5 |
Files
Original bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- Medienwissenschaftliches-Symposium_4_Apprich_2024_Always_.pdf
- Size:
- 694.03 KB
- Format:
- Adobe Portable Document Format
- Description:
- Original PDF with additional cover page.