Data processing: artificial intelligence – Knowledge processing system
Reexamination Certificate
1999-06-09
2003-08-26
Starks, Jr., Wilbert L. (Department: 2121)
Data processing: artificial intelligence
Knowledge processing system
C382S156000, C707S793000
Reexamination Certificate
active
06611825
ABSTRACT:
FIELD OF THE INVENTION
The invention relates generally to text mining, and more specifically to using multidimensional subspaces to represent semantic relationships that exist in a set of documents.
BACKGROUND OF THE INVENTION
Text mining is an extension of the general notion of data mining in the area of free or semi-structured text. Data mining broadly seeks to expose patterns and trends in data, and most data mining techniques are sophisticated methods for analyzing relationships among highly formatted data, i.e., numerical data or data with a relatively small fixed number of possible values. However, much of the knowledge associated with an enterprise consists of textually-expressed information, including free text fields in databases, reports and other documents generated in the company, memos, e-mail, Web sites, and external news articles used by managers, market analysts, and researchers. This data is inaccessible to traditional data mining techniques, because these techniques cannot handle the unstructured or semi-structured nature of free text. Similarly, the analysis task is beyond the capabilities of traditional document management systems and databases. Text mining is a developing field devoted to helping knowledge workers find relationships between individual unstructured or semi-structured text documents and semantic patterns across large collections of such documents.
Research in text mining has its roots in information retrieval. Initial information retrieval work began around 1960, when researchers started to systematically explore methods to match users queries to documents in a database. However, recent advances in computer storage capacity and processing power coupled with massive increases in the amount of text available on-line have resulted in a new emphasis on applying techniques learned from information retrieval to a wider range of text mining problems. Concurrently, text mining has grown from its origins in simple information retrieval systems to encompass additional operations including: information visualization; document classification and clustering; routing and filtering; document summarization; and document cross-referencing. All of the text mining operations listed above share the common need to automatically assess and characterize the similarity between two or more pieces of text. This need is most obvious in information retrieval.
All information retrieval methods depend upon the twin concepts of document and term. A document refers to any body of free or semi-structured text that a user is interested in getting information about in his or her text mining application. This text can be the entire content of a physical or electronic document, an abstract, a paragraph, or even a title. The notion of a document also encompasses text generated from images and graphics or text recovered from audio and video objects. Ideally, a document describes a coherent topic. All documents are represented as collections of terms, and individual terms can appear in multiple documents. Typically, a term is a single word that is used in the text. However, a term can also refer to several words that are commonly used together, for example, “landing gear.” In addition, the terms that represent a piece of text may not appear explicitly in the text; a document's terms may be obtained by applying acronym and abbreviation expansion, word stemming, spelling normalization, thesaurus-based substitutions, or many other techniques. Obtaining the best set of terms for a given document is dependent upon the document or the collection to which the document belongs and the particular goal of the text mining activity.
Once a suitable set of documents and terms have been defined for a text collection, various information retrieval techniques can be applied to the collection. These techniques can be grouped into four broad categories: keyword search methods, natural language understanding methods, probabilistic methods, and vector space methods. Each of these categories is discussed below.
Keyword search methods are currently the most basic and widely used technique in commercial information retrieval systems. The simplest keyword search method retrieves all documents that contain the exact words that are present in the user's query. More advanced keyword search methods trade this cognitive simplicity for more powerful query specification languages, including the ability to specify Boolean operators, proximity operators, fuzzy matching operators, synonym lists, phonetic spellouts, and other term equivalence classes. The advantage of keyword search methods is that they are efficient to implement, very effective for a certain class of queries, and cognitively straightforward. However, keyword search methods have several disadvantages. First, because the technique relies solely on the matching of words, performance is highly dependent on the exact formulation of the query. Second, keyword search methods furnish only crude ways to determine the relevance of a document to the query. The most relevant document may be at the bottom of an enormous list. Third, keyword searching has problems due to the highly ambiguous nature of natural language. Relative to an ideal query response, keyword-based information retrieval systems can both overgenerate and undergenerate. Overgeneration occurs if the query terms have multiple meanings in the document set. Undergeneration occurs if relevant documents in the set happen to use synonyms for the query terms. Finally, keyword searching has limited applicability in general text mining. Keyword searching methods are typically effective only for information retrieval applications, and either do not address or only crudely apply to many of the other text mining operations listed above.
Natural language understanding methods are knowledge-intensive techniques that attempt to parse each sentence in the document collection and generate a semantic interpretation. The individual sentential interpretations can then be composed to yield highly detailed semantic characterizations of documents that can be used to support further text mining operations. Natural language understanding techniques typically require several types of detailed knowledge bases: lexical (information about the grammatical classes and meanings of individual words); syntactic (information about the grammar of expressions); semantic (information about how the expressions refer to the world); and pragmatic (information about the likely intentions and goals of the document author). Given sufficient knowledge bases, the natural language understanding approach to text mining can potentially find levels of detail that cannot be found by other methods. However, the difficulty of constructing and maintaining the required knowledge bases is a major drawback of this method. Knowledge bases of sufficient detail typically require a great deal of time and prior knowledge about the document collection and must be continually updated and checked for consistency. Further, the information in these knowledge bases ranges from fairly general world knowledge to completely domain specific facts and expert knowledge, and can easily involve probabilities, uncertainties, contradictions, and other difficult issues. In addition, even when provided with sufficient knowledge bases, natural language understanding algorithms are extremely slow compared to other text mining approaches. Therefore, natural language understanding methods are typically best suited for text mining applications requiring very detailed information from a specified document or small number of documents, and are not well suited for the analysis of large document collections.
The overall goal of probabilistic information retrieval methods is to estimate the probability that a document is relevant to a query, and to use this estimate to construct a relevance-ranked list of query matches. Probabilistic methods differ in how they model the relationship between documents and queries. The most frequently discussed probabilistic model is the b
Billheimer D. Dean
Booker Andrew James
Condliff Michelle Keim
Greaves Mark Thomas
Holt Fredrick Baden
Christensen O'Connor Johnson & Kindness PLLC
Starks, Jr. Wilbert L.
The Boeing Company
LandOfFree
Method and system for text mining using multidimensional... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for text mining using multidimensional..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for text mining using multidimensional... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3111036