Military intelligence analysts use automated tools to exploit physics-based sensor data to construct a spatio-temporal picture of adversary entities, networks, and behaviors on the battlefield. Traditionally, tools di...
详细信息
Military intelligence analysts use automated tools to exploit physics-based sensor data to construct a spatio-temporal picture of adversary entities, networks, and behaviors on the battlefield. Traditionally, tools did not exploit human generated, textual reports, leaving analysts to manually map dots on the map into meaningful entities using background knowledge about adversary equipment, organization, and activity. Current off-the-shelf text extraction techniques underperform on tactical reports due to unique characteristics of the text. Tactical reports typically feature short sentences with simple grammar, but also tend to include jargon and abbreviations, do not follow grammatical rules, and are likely to have spelling errors. Likewise, named entity recognizers have low recall, because few of the names in reports appear in standard dictionaries. We have developed an entity extraction capability tailored to these challenges, and to the specific needs of analysts, as part of a comprehensive exploitation and fusion system. With fewer cues from syntax, our approach uses semantic constraints to disambiguate syntactic patterns, implemented by a hybrid system that post-processes the output from a standard Natural Language Processing (NLP) engine with our custom semantic pattern analysis. Additional functionality extracts military time and location formats – essential elements that enable downstream fusion of extracted entities with sensor information resulting in a compact and meaningful representation of the battlefield situation.
Rule based systems have achieved success in applications such as information retrieval and Natural Language Processing. However, due to the rigidity of pattern matching, these systems typically require a large number ...
详细信息
Rule based systems have achieved success in applications such as information retrieval and Natural Language Processing. However, due to the rigidity of pattern matching, these systems typically require a large number of rules to adequately cover the variations of expression in unstructured text. Consequently, knowledge engineering for a new domain and knowledge maintenance for a fielded system are labor intensive and expensive. In this paper, we present our research on enhancing a rule-based event coding system by relaxing the rigidity of pattern matching with a technique that formulates and matches patterns of the semantics of words instead of literal words. Our technique pairs literal words with semantic vectors that accumulate word meaning from the context of use of the word found in dictionaries, ontologies, and domain corpora. Our method improves the speed, accuracy, and coverage of the event coding algorithm without additional knowledge engineering effort. Operating on semantics instead of syntax, the improved system eases the workload of human analysts who screen input text for critical events. Our algorithms are based on high-dimensional distributed representations, and their effectiveness and versatility derive from the unintuitive properties of such representations---from the mathematical properties of high-dimensional spaces. Our current implementation encodes words, phrases, and rule patterns as semantic vectors using WordNet, We have started experimental evaluation using a large newswire dataset.
The past decade has seen a renaissance in the development of political event data sets. This has been due to at least three sets of factors. First, there have been technological changes that have reduced the cost of p...
详细信息
The past decade has seen a renaissance in the development of political event data sets. This has been due to at least three sets of factors. First, there have been technological changes that have reduced the cost of producing event data, including the availability of information on the Web, the development of specialized systems for automated coding, and the development of machine-assisted systems that reduce the cost of human coding. Second, event data have become much more elaborate than the original state-centric data sets such as WEIS and COPDAB, with a far greater emphasis on substate and nonstate actors, and in some data sets, the incorporation of geospatial information. Finally, there have been major institutional investments, such as support for a number of Uppsala and PRIO data sets, the DARPA ICEWS Asian and global data sets, and various political violence data sets from the US government. This article will first review the major new contributions, with a focus on those represented in this special issue, discuss some of the open problems in the existing data and finally discuss prospects for future development, including the enhanced use of open-source natural language processing tools, standardizing the coding taxonomies, and prospects for near-real-time coding systems.
Traditional approaches to human information processing tend to deal with perception and action planning in isolation, so that an adequate account of the perception-action interface is still missing. On the perceptual ...
详细信息
Traditional approaches to human information processing tend to deal with perception and action planning in isolation, so that an adequate account of the perception-action interface is still missing. On the perceptual side, the dominant cognitive view largely underestimates, and thus fails to account for, the impact of action-related processes on both the processing of perceptual information and on perceptual learning. On the action side most approaches conceive of action planning as a mere continuation of stimulus processing, thus failing to account for the goal-directedness of even the simplest reaction in an experimental task. We propose a new framework for a more adequate theoretical treatment of perception and action planning, in which perceptual contents and action plans are coded in a common representational medium by feature codes with distal reference. Perceived events (perceptions) and to-be-produced events (actions) are equally represented by integrated, task-tuned networks of feature codes - cognitive structures we call event codes. We give an overview of evidence from a wide variety of empirical domains, such as spatial stimulus-response compatibility, sensorimotor synchronization, and ideomotor action, showing that our main assumptions are well supported by the data.
First, we discuss issues raised with respect to the Theory of event coding (TEC)'s scope, that is, its limitations and possible extensions. Then, we address the issue of specificity, that is, the widespread concer...
详细信息
First, we discuss issues raised with respect to the Theory of event coding (TEC)'s scope, that is, its limitations and possible extensions. Then, we address the issue of specificity, that is, the widespread concern that TEC is too unspecified and, therefore, too vague in a number of important respects. Finally, we elaborate on our views about TEC's relations to other important frameworks and approaches in the field like stages models, ecological approaches, and the two-visual-pathways model.
暂无评论