Matthew Mulholland

Title: Innovations of Matthew Mulholland in Human-Machine Dialog

Introduction

Matthew Mulholland is an accomplished inventor based in Ewing, NJ, known for his contributions to the field of human-machine interaction. He holds three patents that focus on enhancing the effectiveness of dialog applications through advanced machine learning techniques. His work aims to improve the way individuals interact with technology, particularly in educational settings.

Latest Patents

One of Mulholland's latest patents is titled "Ensemble-based machine learning characterization of human-machine dialog." This invention involves receiving data from recordings of individuals interacting with dialog applications. The data is parsed using automated speech recognition to produce text, from which features are extracted. These features are then input into an ensemble of machine learning models, each trained to generate scores that characterize various dialog constructs. The final performance score reflects the conversational proficiency of the individual, providing valuable insights into their interaction with the application.

Another significant patent is the "Automated content feedback generation system for non-native spontaneous speech." This system processes electronic audio files containing spontaneous speech in a non-native language. The spoken words are normalized to eliminate stop words and disfluencies. A trained content scoring model assesses the absence of predefined key points related to the prompt, generating a list of absent key points. This list is displayed in a graphical user interface, along with feedback aimed at improving content completeness.

Career Highlights

Matthew Mulholland is currently employed at Educational Testing Service, where he applies his expertise in machine learning and dialog systems. His work contributes to the development of innovative solutions that enhance educational assessments and learning experiences.

Collaborations

Mulholland collaborates with notable colleagues such as Xinhao Wang and Keelan Evanini, who share his commitment to advancing technology in education. Their combined efforts aim to create more effective tools for learners and educators alike.

Conclusion

Matthew Mulholland's innovative patents and contributions to human-machine dialog demonstrate his dedication to improving technology's role in education. His work not only enhances user interaction but also provides valuable feedback mechanisms for non-native speakers.