Achim Rettinger: Unterschied zwischen den Versionen
Are (Diskussion | Beiträge) |
Cj2486 (Diskussion | Beiträge) |
||
(33 dazwischenliegende Versionen von 6 Benutzern werden nicht angezeigt) | |||
Zeile 2: | Zeile 2: | ||
|Vorname=Achim | |Vorname=Achim | ||
|Nachname=Rettinger | |Nachname=Rettinger | ||
− | |Akademischer Titel= | + | |Akademischer Titel=Privatdozent Dr. rer. nat. |
− | |Forschungsgruppe= | + | |Forschungsgruppe=Web Science |
− | |Stellung= | + | |Stellung=Akademischer Rat |
− | |Ehemaliger= | + | |Ehemaliger=Ja |
− | + | |Email=rettinger@kit.edu | |
− | |Email=rettinger@ | ||
− | |||
|Hinweis DE=Nach Vereinbarung. | |Hinweis DE=Nach Vereinbarung. | ||
− | |Publikationen anzeigen= | + | |Bild=are-passbild-2015-hoch-sw-lowres.jpg |
− | |Vorträge anzeigen= | + | |Info=Achim Rettinger ist als <strong>KIT-Nachwuchsgruppenleiter</strong> am AIFB tätig, wo er das <strong>Adaptive Data Analytics</strong> Team leitet. |
+ | |||
+ | <strong>Research Statement:</strong> | ||
+ | Information retrieval and machine learning approaches are running in the background of most of the applications we use in our daily digital life. The assistance they are providing is manifold, but relies only on a set of core information processing tasks, the most prominent ones being retrieval, classification, clustering and prediction of information. <br> | ||
+ | How content with heterogeneous representations, like text documents in different languages or text and images found online and on social media, can be processed jointly is the focus of this research group. <br> | ||
+ | While the human brain has the ability to integrate disparate multi-sensory information into a coherent percept that benefits from all senses (hearing, seeing,…) current information processing technologies lack this ability. <br> | ||
+ | By combining machine learning with natural language processing and semantic technologies we fuse complementing information from all sources such as text, images and knowledge graphs. This enables cross-modal data analytics and provides a more holistic view than each modality separately. | ||
+ | |||
+ | <strong>Mission:</strong> | ||
+ | Learning of knowledge representations and mapping functions that fuse information from multiple heterogeneous data sources in order to investigating how heterogeneous information interdepends. | ||
+ | |||
+ | Data sources: | ||
+ | <ul> | ||
+ | <li>Text, Social Media Language</li> | ||
+ | <li>Images, Videos, Speech</li> | ||
+ | <li>Knowledge Graphs, Structured information</li> | ||
+ | </ul> | ||
+ | |||
+ | Methods: | ||
+ | <ul> | ||
+ | <li>Latent-variable Models</li> | ||
+ | <li>Knowledge Graph grounding and embedding</li> | ||
+ | <li>Text linking and embedding</li> | ||
+ | </ul> | ||
+ | |||
+ | Applications: | ||
+ | <ul> | ||
+ | <li>Media retrieval, analysis and recommendation</li> | ||
+ | <li>Social Aspects</li> | ||
+ | <li>Healthcare Analytics</li> | ||
+ | </ul> | ||
+ | |Info EN=Achim Rettinger is a <strong>KIT Junior Research Group Leader</strong> at AIFB where he is heading the <strong> Adaptive Data Analytics</strong> team. | ||
+ | |||
+ | <strong>Research Statement:</strong> | ||
+ | Information retrieval and machine learning approaches are running in the background of most of the applications we use in our daily digital life. The assistance they are providing is manifold, but relies only on a set of core information processing tasks, the most prominent ones being retrieval, classification, clustering and prediction of information. | ||
+ | How content with heterogeneous representations, like text documents in different languages or text and images found online and on social media, can be processed jointly is the focus of this research group. <br> | ||
+ | While the human brain has the ability to integrate disparate multi-sensory information into a coherent percept that benefits from all senses (hearing, seeing,…) current information processing technologies lack this ability. <br> | ||
+ | By combining machine learning with natural language processing and semantic technologies we fuse complementing information from all sources such as text, images and knowledge graphs. This enables cross-modal data analytics and provides a more holistic view than each modality separately. | ||
+ | |||
+ | <strong>Mission:</strong> | ||
+ | Learning of knowledge representations and mapping functions that fuse information from multiple heterogeneous data sources in order to investigating how heterogeneous information interdepends. | ||
+ | |||
+ | Data sources: | ||
+ | <ul> | ||
+ | <li>Text, Social Media Language</li> | ||
+ | <li>Images, Videos, Speech</li> | ||
+ | <li>Knowledge Graphs, Structured information</li> | ||
+ | </ul> | ||
+ | |||
+ | Methods: | ||
+ | <ul> | ||
+ | <li>Latent-variable Models</li> | ||
+ | <li>Knowledge Graph grounding and embedding</li> | ||
+ | <li>Text linking and embedding</li> | ||
+ | </ul> | ||
+ | |||
+ | Applications: | ||
+ | <ul> | ||
+ | <li>Media retrieval, analysis and recommendation</li> | ||
+ | <li>Social Aspects</li> | ||
+ | <li>Healthcare Analytics</li> | ||
+ | </ul> | ||
+ | |Publikationen anzeigen=Ja | ||
+ | |Vorträge anzeigen=Nein | ||
|Organisation=AIFB, KIT | |Organisation=AIFB, KIT | ||
+ | |Abschlussarbeiten anzeigen=Nein | ||
+ | }} | ||
+ | {{Forschungsgebiet Auswahl | ||
+ | |Forschungsgebiet=Data Mining | ||
+ | }} | ||
+ | {{Forschungsgebiet Auswahl | ||
+ | |Forschungsgebiet=Informationsextraktion | ||
+ | }} | ||
+ | {{Forschungsgebiet Auswahl | ||
+ | |Forschungsgebiet=Knowledge Discovery | ||
+ | }} | ||
+ | {{Forschungsgebiet Auswahl | ||
+ | |Forschungsgebiet=Ontology Learning | ||
+ | }} | ||
+ | {{Forschungsgebiet Auswahl | ||
+ | |Forschungsgebiet=Maschinelles Lernen | ||
+ | }} | ||
+ | {{Forschungsgebiet Auswahl | ||
+ | |Forschungsgebiet=Mensch-Maschine-Systeme | ||
+ | }} | ||
+ | {{Forschungsgebiet Auswahl | ||
+ | |Forschungsgebiet=Text Mining | ||
}} | }} |
Aktuelle Version vom 23. September 2019, 12:47 Uhr
Privatdozent Dr. rer. nat. Achim Rettinger |
||
Ehemaliges Mitglied
|
|
Achim Rettinger ist als KIT-Nachwuchsgruppenleiter am AIFB tätig, wo er das Adaptive Data Analytics Team leitet.
Research Statement:
Information retrieval and machine learning approaches are running in the background of most of the applications we use in our daily digital life. The assistance they are providing is manifold, but relies only on a set of core information processing tasks, the most prominent ones being retrieval, classification, clustering and prediction of information.
How content with heterogeneous representations, like text documents in different languages or text and images found online and on social media, can be processed jointly is the focus of this research group.
While the human brain has the ability to integrate disparate multi-sensory information into a coherent percept that benefits from all senses (hearing, seeing,…) current information processing technologies lack this ability.
By combining machine learning with natural language processing and semantic technologies we fuse complementing information from all sources such as text, images and knowledge graphs. This enables cross-modal data analytics and provides a more holistic view than each modality separately.
Mission: Learning of knowledge representations and mapping functions that fuse information from multiple heterogeneous data sources in order to investigating how heterogeneous information interdepends.
Data sources:
- Text, Social Media Language
- Images, Videos, Speech
- Knowledge Graphs, Structured information
Methods:
- Latent-variable Models
- Knowledge Graph grounding and embedding
- Text linking and embedding
Applications:
- Media retrieval, analysis and recommendation
- Social Aspects
- Healthcare Analytics