Julian Cerruti

Buenos Aires, Provincia de Buenos Aires, Argentina
1 mil seguidores Más de 500 os

Unirse para ver el perfil

Acerca de

I am a software engineering leader with a focus on robotics and large scale systems…

Actividad

Unirse para ver toda la actividad

Experiencia y educación

  • InOrbit.AI

Mira la experiencia completa de Julian

Mira su cargo, antigüedad y más

o

Al hacer clic en «Continuar» para unirte o iniciar sesión, aceptas las Condiciones de uso, la Política de privacidad y la Política de cookies de LinkedIn.

Publicaciones

  • Function words as a domain independent tool for social role inference

    Universidad Nacional de Córdoba - Facultad de Matemática, Astronomía y Física

    I investigated the automated role inference task in virtual communities under the scope of IBM Communities. I consider this task as a classication problem of two explicit roles: community owners and as they represent the leader and non-leader roles. I began with a Human Performance test to measure how well humans can infer leadership roles from the content posted obtaining a 76.25% of accuracy. Then, I reproduced three standard approaches in Information Retrieval. I trained six…

    I investigated the automated role inference task in virtual communities under the scope of IBM Communities. I consider this task as a classication problem of two explicit roles: community owners and as they represent the leader and non-leader roles. I began with a Human Performance test to measure how well humans can infer leadership roles from the content posted obtaining a 76.25% of accuracy. Then, I reproduced three standard approaches in Information Retrieval. I trained six automated classiers to learn which words are relevant for each role so they can be used as domain dependent features for the classication, obtaining 75.26% of accuracy. Finally, our contribution was to propose three domain independent approaches: two of them based in the sociolinguistic theory of politeness, and the third one based in Pennebaker's theory of function words as social carriers. I trained the same six classiers to test both theories obtaining an accuracy of 69.12% for the Function Words approach, and nding evidence that leaders make use of positive politeness strategies. I conclude showing that the results achieved by the most common approaches are domain dependent while the domain independent performance of function words is good enough to predict social roles.

    Otros autores
    Ver publicación

Idiomas

  • Spanish

    -

Recomendaciones recibidas

1 persona ha recomendado a Julian

Unirse para verlo

Más actividad de Julian

Ver el perfil completo de Julian

  • Descubrir a quién conocéis en común
  • Conseguir una presentación
  • ar con Julian directamente
Unirse para ver el perfil completo

Perfiles similares

Otras personas con el nombre de Julian Cerruti

Hay 6 personas más con el nombre de Julian Cerruti en LinkedIn

Ver a otras personas con el nombre de Julian Cerruti

Añade nuevas aptitudes con estos cursos

Ver todos los cursos