Goran Glavaš

Title: Goran Glavaš: Innovator in Text Segmentation Technology

Introduction

Goran Glavaš is a notable inventor based in Heidelberg, Germany. He has made significant contributions to the field of text segmentation through his innovative patent. His work focuses on enhancing the processing of unstructured text data, which is increasingly relevant in today's data-driven world.

Latest Patents

Goran Glavaš holds a patent for "Training and domain adaptation for supervised text segmentation." This patent involves receiving unstructured text data that includes a sequence of sentences. The data is tokenized into a plurality of tokens, which are then segmented using a hierarchical transformer network model. This model comprises a token transformer, a sentence transformer, and a segmentation classifier. The token transformer contextualizes tokens within sentences to yield sentence embeddings. The sentence transformer contextualizes sentence representations based on these embeddings. Finally, the segmentation classifier predicts segments of the received data based on the contextualized sentence representations. This innovative approach allows for more accurate segmentation of text data, which can be crucial for various applications.

Career Highlights

Goran Glavaš is currently employed at Educational Testing Service, where he applies his expertise in text processing and machine learning. His work at this organization allows him to contribute to advancements in educational assessments and testing methodologies.

Collaborations

Goran collaborates with Swapna Somasundaran, who is also involved in the field of educational testing and technology. Their partnership enhances the development of innovative solutions in text segmentation and analysis.

Conclusion

Goran Glavaš is a pioneering inventor whose work in text segmentation technology is shaping the future of data processing. His contributions are vital for improving the accuracy and efficiency of handling unstructured text data.