Yuki User

- Updated Named Entity Recognition (NER) tool for Tibetan, added to project UI;
- The UI is accessible via an entry page and via a public repository
- Another version of NER code incorporated in the UI and package;
- All necessary documentation and repository postings.

- Identified and catalogued named entities — with a focus on historical persons and anonymous references to them — across 126 Tibetan-language works, working from XML transcription drafts and manuscript reproductions.

Identified and catalogued named entities, focusing on historical persons and anonymous references, across 126 Tibetan-language works from XML transcription drafts and manuscript reproductions.

- Identified and catalogued named entities — with a focus on historical persons and anonymous references to them — across 126 Tibetan-language works, working from XML transcription drafts and manuscript reproductions.

- Development of NER for modern Tibetan, using Claude Anthropic to develop training data and spaCy to carry out data extraction
- Export of NER data extracted from the Project corpus into a user-friendly format such as a .csv file, including a reference to the source for each data item and preserving significant co-occurrences (such as name+position or name+associated slogan) where possible
- Cons

https://research.uni-leipzig.de/diverge/
The followings are my tasks:
- NER (Named Entity Recognition): Collect available lists of named entities in Tibetan, including personal and place names.
- spaCy training data: Produce a script to create training data in modern Tibetan for a Tibetan language model for spaCy. (Usually one needs a dataset whose entries are manually annotated, but we would like

https://research.uni-leipzig.de/diverge/
The followings are my tasks:
- NER (Named Entity Recognition): Collect available lists of named entities in Tibetan, including personal and place names.
- spaCy training data: Produce a script to create training data in modern Tibetan for a Tibetan language model for spaCy. (Usually one needs a dataset whose entries are manually annotated, but we would like

- Development of a web-based engineering platform to simplify the electronics development process.
- Development of a method for AI-based processing of the reference designs, which enables automatic generation of the digital modules.
- Try out tools, which are potentially applicable to our project, to create training data and to evaluate test results.

- Development of a web-based engineering platform to simplify the electronics development process.
- Development of a method for AI-based processing of the reference designs, which enables automatic generation of the digital modules.
- Try out tools, which are potentially applicable to our project, to create training data and to evaluate test results.

- Shannon's information theory
- Automatic keywords extraction
- Automatic summarization
- Evaluation method for summarization
- information retrieval

- Shannon's information theory
- Automatic keywords extraction
- Automatic summarization
- Evaluation method for summarization
- information retrieval

- Data Analysis using R, Clustering (k-means)
- Maintenance of a system using SAS

- Data Analysis using R, Clustering (k-means)
- Maintenance of a system using SAS

Without getting degree

Without getting degree

Doctor of Philosophy (PhD) in Indian Philosophy at Hokkaido University, without receiving the degree.

Master of Philosophy and Cultural Sciences in Indian Philosophy at Hokkaido University.

Doctor of Philosophy (PhD) in Indology at Leipzig University.