Conference Paper

Sentence Induced Transformations in Conceptual Spaces

DINFO, Univ. of Palermo, Palermo
DOI: 10.1109/ICSC.2008.74 Conference: Semantic Computing, 2008 IEEE International Conference on
Source: IEEE Xplore

ABSTRACT The proposed work illustrates how "primitive concepts" can be automatically induced from a text corpus. The primitive concepts are identified by the orthonormal axis of a "conceptual" space induced by a methodology inspired tothe latent semantic analysis approach. The methodology represents a natural language sentence by means of a set of rotations of an orthonormal basis in the "conceptual"space. The rotations, triggered by the sequence of words composing the sentence and realized by means of geometric algebra rotors, allow to highlight "conceptual" relations that can arise among the primitive concepts.

0 Bookmarks
 · 
96 Views
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To enable intention-based search, computers have to compare the meaning of user’s search intentions and data items being searched. Traditional information retrieval systems fail to adequately represent, compare and discern composite and finer meanings. This research proposes to develop techniques to represent and compare composite meanings. The proposed contribution are: (1) an algebraic theory, design of an data structure and related algorithms for representing hierarchically composed meaning as a tensor which is amenable to efficient semantic similarity comparison; and (2) techniques to construct a scalable and self-organizing meaning based index system that can seamlessly federate and integrate multiple distributed repositories.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we present a fine grained parallel architecture that performs meaning comparison using vector cosine similarity (dot product). Meaning comparison assigns a similarity value to two objects (e.g. text documents) based on how similar their meanings (represented as two vectors) are to each other. The novelty of our design is the fine grained parallelism which is not exploited in available hardware based dot product processor designs and can not be achieved in traditional server class processors like the Intel Xeon. We compare the performance of our design against that of available hardware based dot product processors as well a server class processor using optimum software code performing the same computation. We show that our hardware design can achieve a speedup of 62,000 times compared to an available hardware design and a speedup of 8866 times with 33% (1.5 times) less power consumption, compared to software code running on Intel Xeon processor for 1024 basis vectors. Our design can significantly reduce the amount of servers required for similarity comparison in a distributed search engine. Thus it can enable reduction in energy consumption, investment, operational costs and floor area in search engine data centers. This design can also be deployed for other applications which require fast dot product computation.
    Parallel & Distributed Processing (IPDPS), 2010 IEEE International Symposium on; 05/2010
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we present a fine grained parallel architecture that performs meaning comparison using vector cosine similarity (dot product). Meaning comparison assigns a similarity value to two objects (e.g. text documents) based on how similar their meanings (represented as two vectors) are to each other. The novelty of our design is the fine grained parallelism which is not exploited in available hardware based dot product processor designs and can not be achieved in traditional server class processors like the Intel Xeon. We compare the performance of our design against that of available hardware based dot product processors as well a server class processor using optimum software code performing the same computation. We show that our hardware design can achieve a speedup of 62,000 times compared to an available hardware design and a speedup of 8866 times with 33% (1.5 times) less power consumption, compared to software code running on Intel Xeon processor for 1024 basis vectors. Our design can significantly reduce the amount of servers required for similarity comparison in a distributed search engine. Thus it can enable reduction in energy consumption, investment, operational costs and floor area in search engine data centers. This design can also be deployed for other applications which require fast dot product computation.
    24th IEEE International Parallel and Distributed Processing Symposium (IEEE IPDPS), Atlanta, Apr 2010.; 01/2010