Abstract
Soft information fusion, fusing information from natural language messages with other soft information and with information from physical sensors is facilitated by representing the information in the messages as a formally defined propositional graph that abides by the uniqueness principle—the principle that every entity or event that is mentioned in the message is represented by a unique node in the graph, or, at worst, by several nodes connected by co-referentiality relations. To further facilitate information fusion, information from the message is enhanced with relevant information from background knowledge sources. What knowledge is relevant is determined by also representing the background knowledge as a propositional graph, embedding the knowledge graph from the messages into the background knowledge graph using the uniqueness principle to fuse a message graph node with a background knowledge graph node, and then using spreading activation to find subgraphs of the background knowledge graph. This combination of the message graph with the retrieved subgraphs is considered the “relevant information.” In this chapter, we discuss, evaluate, and compare two techniques for spreading activation.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Node wft4! has been left uncollapsed in preparation for Fig. 14.3.
- 2.
More appropriately, this can be viewed as constraining the memory encoding process of some agent, though a knowledge engineer is performing the encoding process.
- 3.
A value of 1.0 is used for the initial activation value since it is recommended in the Texai approach, and will always be greater than or equal to the activation threshold.
- 4.
The activation threshold was 0.5 and decay was 0.9 (c.f., Sect. 14.5.1.1).
- 5.
These rules are provisional ones created for testing purposes, and are probably not those a SME would come up with for this domain.
- 6.
The actual representation of this reasoning rule is much more specific.
- 7.
ACT-R representations typically use smaller knowledge bases than those in large-scale systems, like Cyc [36], that require information retrieval techniques. As such, the activation calculation used in ACT-R gives a ranking to all the information in declarative memory and then selects the best ranked results as a match.
- 8.
The ACT-R specification [29] recommends a value of 0.5 for d after numerous tests, but this was for the retrieval of one chunk and may be different for using the spreading activation algorithm for information retrieval.
- 9.
The equation provided in [29] calculates the associative strength as \(S_{ji} \approx ln(prob(i|j)/prob(i))\), but provides no specification for calculating the probabilities in a propositional graph.
- 10.
At the time of this study, Tractor was in its infancy, and thus the messages had to be manually translated into SNePS 3 propositional graphs. This limited the number of examples we could use in the evaluation.
- 11.
The SNePS 3 KRR system, background knowledge sources and means of loading them into SNePS 3, message representations, and code for evaluating the algorithms is available at http://www.cse.buffalo.edu/~mwk3/Papers/evaluation.html.
- 12.
An origin set for a proposition is the set of propositions used in the derivation of that proposition. Origin sets originate from relevance logic proof theory [40].
References
G.A. Gross, R. Nagi, K. Sambhoos, D.R. Schlegel, S.C. Shapiro, G. Tauer, Towards hard + soft data fusion: processing architecture and implementation for the joint fusion and analysis of hard and soft intelligence data, in Proceedings of the 15th International Conference on Information Fusion (Fusion 2012) (ISIF, 2012), pp. 955–962
M. Prentice, M. Kandefer, S.C. Shapiro, Tractor: a framework for soft information fusion, in Proceedings of the 13th International Conference on Information Fusion (Fusion2010) (2010), pp. Th3.2.2
S.C. Shapiro, D.R. Schlegel, Natural language understanding for soft information fusion, in Proceedings of the 16th International Conference on Information Fusion (Fusion 2013) (ISIF, 2013), 9 p (unpaginated)
A.B. Poore, S. Lu, B.J. Suchomel, Data association using multiple frame assignments, in Handbook of Multisensor Data Fusion (chapter 13), 2nd ed. ed. by M. Liggins, D. Hall, J. Llinas (CRC Press, 2009), pp. 299–318
J.L. Graham, A new synthetic dataset for evaluating soft and hard fusion algorithms, in Proceedings of the SPIE Defense, Security, and Sensing Symposium: Defense Transformation and Net-Centric Systems 2011 (2011), pp. 25–29
D.R. Schlegel, S.C. Shapiro, Visually interacting with a knowledge base using frames, logic, and propositional graphs, in Graph Structures for Knowledge Representation and Reasoning, vol. 7205, Lecture Notes in Artificial Intelligence, ed. by M. Croitoru, S. Rudolph, N. Wilson, J. Howse, O. Corby (Springer, Berlin, 2012), pp. 188–207
S.C. Shapiro, Belief spaces as sets of propositions. J. Exp. Theor. Artif. Intell. (JETAI), 5(2 and 3):225–235 (1993, Apr–Sept)
S.C. Shapiro, W.J. Rapaport, The SNePS family. Comput. Math. Appl. 23(2–5):243–275 (1992, Jan–Mar) Reprinted in [24, pp. 243–275]
S.C. Shapiro, Symmetric relations, intensional individuals, and variable binding. Proc. IEEE 74(10), 1354–1363 (1986)
A.S. Maida, S.C. Shapiro, Intensional concepts in propositional semantic networks. Cogn. Sci. 6(4):291–330 (1982, Oct–Dec). Reprinted in [7, pp. 170–189]
S.C. Shapiro, Cables, paths and “subconscious” reasoning in propositional semantic networks, in Principles of Semantic Networks: Explorations in the Representation of Knowledge, ed. by J. Sowa (Morgan Kaufmann, Los Altos, CA, 1991), pp. 137–156
A.N. Steinberg, G. Rogova, Situation and context in data fusion and natural language understanding, in Proceedings of the 11th International Conference on Information Fusion (Fusion2008) (IEEE, 2008, June), pp. 1–8
S.C. Shapiro, An introduction to SNePS 3, in Conceptual Structures: Logical, Linguistic, and Computational Issues, vol. 1867, Lecture Notes in Artificial Intelligence, ed. by B. Ganter, G.W. Mineau (Springer, Berlin, 2000), pp. 510–524
N.A. Bradley, M.D. Dunlop, Toward a multidisciplinary model of context to support contextaware computing. Hum. Comput. Interact. 20(4), 403–446 (2005)
D. Godden, A. Baddeley, Contextdependent memory in two natural environments: on land and underwater. Br. J. Psychol. 66, 325–332 (1975)
S. Smith, E. Vela, Environmental context-dependent memory. Psychon. Bull. Rev. 8(2), 203–220 (2001)
S. Smith, Remembering in and out of context. J. Exp. Psychol. Hum. Learn. Memory 5(5), 460–471 (1979)
A.K. Dey, Understanding and using context. Pers. Ubiquit. Comput. 5(1), 4–7 (2001)
G. Chen, D. Kotz, A survey of context-aware mobile computing research. Technical Report TR2000–381, Department of Computer Science, Dartmouth College, Hanover, NH, November 2000
J. McCarthy, Generality in artificial intelligence. Commun. ACM 30(12), 1030–1035 (1987)
M. Benerecetti, P. Bouquet, C. Ghidini, Contextual reasoning distilled. JETAI 12(3), 279–305 (2000)
P. Bouquet, C. Ghidini, F.O. Giunchiglia, E. Blanzieri, Theories and uses of context in knowledge representation and reasoning. J. Pragmat. 35(3), 403–446 (2003)
S. Buvač, Quantificational logic of context, in Proceedings of the Thirteenth National Conference on Artificial Intelligence (AAAI-96), Menlo Park, CA (AAAI Press, 1996), pp. 600–606
P. Brézillon, Context in artificial intelligence: I. A survey of the literature. Comput. Artif. Intell. 18(4), 321–340 (1999)
P. Brézillon, Context in problem solving. Knowl. Eng. Rev. 14(1), 47–80 (1999)
P. Brézillon, Context in artificial intelligence: II. Key elements of contexts. Comput. Artif. Intell. 18(5), 425–446 (1999)
D. Lenat, The dimensions of context-space. Technical report, Cycorp, October 1998
R.P. Arritt, R.M. Turner, Situation assessment for autonomous underwater vehicles using a priori contextual knowledge, in Proceedings of the Thirteenth International Symposium on Unmanned Untethered Submersible Technology (UUST) (2003)
J. R. Anderson. Human associative memory. In How Can the Human Mind Occur in the Physical Universe? (Oxford University Press, New York, 2007), pp. 91–134
A.M. Collins, E.F. Loftus, A spreading activation theory of semantic processing. Psychol. Rev. 82(6), 407–428 (1975)
F. Crestani, Application of spreading activation techniques in information retrieval. Artif. Intell. Rev. 11(6), 453–482 (1997)
S.L. Reed, Texai (2010). http://sourceforge.net/projects/texai/
J.R. Anderson, Cognitive architecture, in How Can the Human Mind Occur in the Physical Universe?, pages 3–43. Oxford University Press, New York, 2007
M.B. Howes, Long-term memory: ongoing research, in Human Memory: Structures and Images (SAGE Publications, Thousand Oaks, 2007)
D. Bothell. ACT-R 6.0 Reference Manual (2010)
D.B. Lenat, Cyc: a large-scale investment in knowledge infrastructure. Commun. ACM 38(11), 33–38 (1995)
National Geospatial-Intelligence Agency. NGA GEOnet Names Server (2014). http://earth-info.nga.mil/gns/html/
C.J. van Rijsbergen, Information Retrieval (Butterworths, London, second edition, 1979)
M. Kandefer, S.C. Shapiro, An F-measure for context-based information retrieval, in Commonsense 2009: Proceedings of the Ninth International Symposium on Logical Formalizations of Commonsense Reasoning, ed. by G. Lakemeyer, L. Morgenstern, M.-A. Williams (The Fields Institute, Toronto, CA, 2009), pp. 79–84
S.C. Shapiro, Relevance logic in computer science, in Entailment, volume II, ed. by A.R. Anderson, N.D. Belnap Jr, M. Dunn (Princeton University Press, Princeton, 1992), pp. 553–563
M. Kandefer and S. C. Shapiro. Evaluating spreading activation for soft information fusion, in Proceedings of the 14th International Conference on Information Fusion (Fusion 2011), pages 498–505. ISIF, 2011
M.W. Kandefer, S.C. Shapiro, A categorization of contextual constraints, in Biologically Inspired Cognitive Architectures: Papers from the 2008 AAAI Fall Symposium, ed. by A. Samsonovich (AAAI Press, Menlo Park, CA, 2008), pp. 88–93
R.J. Brachman, H.J. Levesque (eds.), Readings in knowledge representation (Morgan Kaufmann, San Mateo, 1985)
F. Lehmann (ed.), Semantic Networks in Artificial Intelligence (Pergamon Press, Oxford, 1992)
Acknowledgements
This work was supported in part by the Office of Naval Research under contract N00173-08-C-4004, and by a Multidisciplinary University Research Initiative (MURI) grant (Number W911NF-09-1-0392) for “Unified Research on Network-based Hard/Soft Information Fusion ,” issued by the US Army Research Office (ARO) under the program management of Dr. John Lavery. The work describe here was done while both authors were in the Department of Computer Science and Engineering, University at Buffalo, Buffalo, NY. Parts of this paper were taken from [39, 41, 42].
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland (outside the USA)
About this chapter
Cite this chapter
Kandefer, M., Shapiro, S.C. (2016). Context Relevance for Text Analysis and Enhancement for Soft Information Fusion. In: Snidaro, L., García, J., Llinas, J., Blasch, E. (eds) Context-Enhanced Information Fusion. Advances in Computer Vision and Pattern Recognition. Springer, Cham. https://doi.org/10.1007/978-3-319-28971-7_14
Download citation
DOI: https://doi.org/10.1007/978-3-319-28971-7_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-28969-4
Online ISBN: 978-3-319-28971-7
eBook Packages: Computer ScienceComputer Science (R0)