Literatur zur Informationserschließung

Diese Datenbank enthält über 40.000 Dokumente zu Themen aus den Bereichen Formalerschließung – Inhaltserschließung – Information Retrieval.

© 2015 W. Gödert, TH Köln, Institut für Informationswissenschaft / Powered by litecat, BIS Oldenburg (Stand: 04. Juni 2021)

Suchergebnisse

• 1
  Ma, N. ; Zheng, H.T. ; Xiao, X.: ¬An ontology-based latent semantic indexing approach using long short-term memory networks.
  In: Web and Big Data: First International Joint Conference, APWeb-WAIM 2017, Beijing, China, July 7-9, 2017, Proceedings, Part I. Eds.: L. Chen et al. Cham : Springer, 2017. S.185-199.
  (Lecture notes in computer science; vol.10366)
  
  Abstract: Nowadays, online data shows an astonishing increase and the issue of semantic indexing remains an open question. Ontologies and knowledge bases have been widely used to optimize performance. However, researchers are placing increased emphasis on internal relations of ontologies but neglect latent semantic relations between ontologies and documents. They generally annotate instances mentioned in documents, which are related to concepts in ontologies. In this paper, we propose an Ontology-based Latent Semantic Indexing approach utilizing Long Short-Term Memory networks (LSTM-OLSI). We utilize an importance-aware topic model to extract document-level semantic features and leverage ontologies to extract word-level contextual features. Then we encode the above two levels of features and match their embedding vectors utilizing LSTM networks. Finally, the experimental results reveal that LSTM-OLSI outperforms existing techniques and demonstrates deep comprehension of instances and articles.
  
  Inhalt: Vgl.: https://link.springer.com/chapter/10.1007/978-3-319-63579-8_15. DOI: https://doi.org/10.1007/978-3-319-63579-8_15.
  
  Themenfeld: Wissensrepräsentation ; Semantisches Umfeld in Indexierung u. Retrieval ; Automatisches Indexieren
  Objekt: Latent Semantic Indexing
  
  
• 2
  Grün, S.: Mehrwortbegriffe und Latent Semantic Analysis : Bewertung automatisch extrahierter Mehrwortgruppen mit LSA.
  Düsseldorf : Heinrich-Heine-Universität / Philosophische Fakultät / Institut für Sprache und Information, 2017. 67 S.
  
  Abstract: Die vorliegende Studie untersucht das Potenzial von Mehrwortbegriffen für das Information Retrieval. Zielsetzung der Arbeit ist es, intellektuell positiv bewertete Kandidaten mithilfe des Latent Semantic Analysis (LSA) Verfahren höher zu gewichten, als negativ bewertete Kandidaten. Die positiven Kandidaten sollen demnach bei einem Ranking im Information Retrieval bevorzugt werden. Als Kollektion wurde eine Version der sozialwissenschaftlichen GIRT-Datenbank (German Indexing and Retrieval Testdatabase) eingesetzt. Um Kandidaten für Mehrwortbegriffe zu identifizieren wurde die automatische Indexierung Lingo verwendet. Die notwendigen Kernfunktionalitäten waren Lemmatisierung, Identifizierung von Komposita, algorithmische Mehrworterkennung sowie Gewichtung von Indextermen durch das LSA-Modell. Die durch Lingo erkannten und LSAgewichteten Mehrwortkandidaten wurden evaluiert. Zuerst wurde dazu eine intellektuelle Auswahl von positiven und negativen Mehrwortkandidaten vorgenommen. Im zweiten Schritt der Evaluierung erfolgte die Berechnung der Ausbeute, um den Anteil der positiven Mehrwortkandidaten zu erhalten. Im letzten Schritt der Evaluierung wurde auf der Basis der R-Precision berechnet, wie viele positiv bewerteten Mehrwortkandidaten es an der Stelle k des Rankings geschafft haben. Die Ausbeute der positiven Mehrwortkandidaten lag bei durchschnittlich ca. 39%, während die R-Precision einen Durchschnittswert von 54% erzielte. Das LSA-Modell erzielt ein ambivalentes Ergebnis mit positiver Tendenz.
  
  Anmerkung: Masterarbeit, Studiengang Informationswissenschaft und Sprachtechnologie, Institut für Sprache und Information, Philosophische Fakultät, Heinrich-Heine-Universität Düsseldorf
  
  Themenfeld: Automatisches Indexieren
  Objekt: Lingo ; Latent Semantic Indexing ; GIRT
  
  
• 3
  Zhu, W.Z. ; Allen, R.B.: Document clustering using the LSI subspace signature model.
  In: Journal of the American Society for Information Science and Technology. 64(2013) no.4, S.844-860.
  
  Abstract: We describe the latent semantic indexing subspace signature model (LSISSM) for semantic content representation of unstructured text. Grounded on singular value decomposition, the model represents terms and documents by the distribution signatures of their statistical contribution across the top-ranking latent concept dimensions. LSISSM matches term signatures with document signatures according to their mapping coherence between latent semantic indexing (LSI) term subspace and LSI document subspace. LSISSM does feature reduction and finds a low-rank approximation of scalable and sparse term-document matrices. Experiments demonstrate that this approach significantly improves the performance of major clustering algorithms such as standard K-means and self-organizing maps compared with the vector space model and the traditional LSI model. The unique contribution ranking mechanism in LSISSM also improves the initialization of standard K-means compared with random seeding procedure, which sometimes causes low efficiency and effectiveness of clustering. A two-stage initialization strategy based on LSISSM significantly reduces the running time of standard K-means procedures.
  
  Themenfeld: Automatisches Klassifizieren
  Objekt: Latent semantic indexing
  
  
• 4
  Sojka, P. ; Lee, M. ; Rehurek, R. ; Hatlapatka, R. ; Kucbel, M. ; Bouche, T. ; Goutorbe, C. ; Anghelache, R. ; Wojciechowski, K.: Toolset for entity and semantic associations : Final Release.Revision: 1.0 as of 8th February 2013.
  In: https://wiki.eudml.eu/eudml-w/images/D8.4-v1.0.pdf.
  
  Abstract: In this document we describe the final release of the toolset for entity and semantic associations, integrating two versions (language dependent and language independent) of Unsupervised Document Similarity implemented by MU (using gensim tool) and Citation Indexing, Resolution and Matching (UJF/CMD). We give a brief description of tools, the rationale behind decisions made, and provide elementary evaluation. Tools are integrated in the main project result, EuDML website, and they deliver the needed functionality for exploratory searching and browsing the collected documents. EuDML users and content providers thus benefit from millions of algorithmically generated similarity and citation links, developed using state of the art machine learning and matching methods.
  
  Inhalt: Vgl. auch: https://is.muni.cz/repo/1076213/en/Lee-Sojka-Rehurek-Bolikowski/Toolset-for-Entity-and-Semantic-Associations-Initial-Release-Deliverable-82-of-project-EuDML?lang=en.
  
  Themenfeld: Automatisches Klassifizieren
  Wissenschaftsfach: Mathematik
  Objekt: GENSIM ; Latent Semantic Indexing ; Zentralblatt für Mathematik
  
  
• 5
  Kumar, C.A. ; Radvansky, M. ; Annapurna, J.: Analysis of Vector Space Model, Latent Semantic Indexing and Formal Concept Analysis for information retrieval.
  In: Cybernetics and information technologies. 12(2012) no.1, S.34-48.
  
  Abstract: Latent Semantic Indexing (LSI), a variant of classical Vector Space Model (VSM), is an Information Retrieval (IR) model that attempts to capture the latent semantic relationship between the data items. Mathematical lattices, under the framework of Formal Concept Analysis (FCA), represent conceptual hierarchies in data and retrieve the information. However both LSI and FCA uses the data represented in form of matrices. The objective of this paper is to systematically analyze VSM, LSI and FCA for the task of IR using the standard and real life datasets.
  
  Themenfeld: Formale Begriffsanalyse
  Objekt: Latent semantic indexing
  
  
• 6
  Wicaksana, I.W.S. ; Wahyudi, B.: Comparison Latent Semantic and WordNet approach for semantic similarity calculation.
  In: http://arxiv.org/find/all/1/all:+EXACT+semantic_interoperability/0/1/0/all/0/1. [arXiv:1105.1406].
  
  Abstract: Information exchange among many sources in Internet is more autonomous, dynamic and free. The situation drive difference view of concepts among sources. For example, word 'bank' has meaning as economic institution for economy domain, but for ecology domain it will be defined as slope of river or lake. In this paper, we will evaluate latent semantic and WordNet approach to calculate semantic similarity. The evaluation will be run for some concepts from different domain with reference by expert or human. Result of the evaluation can provide a contribution for mapping of concept, query rewriting, interoperability, etc.
  
  Themenfeld: Semantische Interoperabilität
  Objekt: Latent semantic indexing ; WordNet
  
  
• 7
  Zhang, W. ; Yoshida, T. ; Tang, X.: ¬A comparative study of TF*IDF, LSI and multi-words for text classification.
  In: Expert-systems with applications. 38(2011) no.3, S.2758-2765.
  
  Abstract: One of the main themes in text mining is text representation, which is fundamental and indispensable for text-based intellegent information processing. Generally, text representation inludes two tasks: indexing and weighting. This paper has comparatively studied TF*IDF, LSI and multi-word for text representation. We used a Chinese and an English document collection to respectively evaluate the three methods in information retreival and text categorization. Experimental results have demonstrated that in text categorization, LSI has better performance than other methods in both document collections. Also, LSI has produced the best performance in retrieving English documents. This outcome has shown that LSI has both favorable semantic and statistical quality and is different with the claim that LSI can not produce discriminative power for indexing.
  
  Inhalt: Vgl. unter: http://www.sciencedirect.com/science/article/pii/S0957417410008626.
  
  Themenfeld: Semantisches Umfeld in Indexierung u. Retrieval ; Retrievalalgorithmen
  Objekt: Latent Semantic Indexing
  
  
• 8
  Li, D. ; Kwong, C.-P.: Understanding latent semantic indexing : a topological structure analysis using Q-analysis.
  In: Journal of the American Society for Information Science and Technology. 61(2010) no.3, S.592-608.
  
  Abstract: The method of latent semantic indexing (LSI) is well-known for tackling the synonymy and polysemy problems in information retrieval; however, its performance can be very different for various datasets, and the questions of what characteristics of a dataset and why these characteristics contribute to this difference have not been fully understood. In this article, we propose that the mathematical structure of simplexes can be attached to a term-document matrix in the vector space model (VSM) for information retrieval. The Q-analysis devised by R.H. Atkin ([1974]) may then be applied to effect an analysis of the topological structure of the simplexes and their corresponding dataset. Experimental results of this analysis reveal that there is a correlation between the effectiveness of LSI and the topological structure of the dataset. By using the information obtained from the topological analysis, we develop a new method to explore the semantic information in a dataset. Experimental results show that our method can enhance the performance of VSM for datasets over which LSI is not effective.
  
  Objekt: Latent Semantic Indexing ; Q-analysis
  
  
• 9
  Rehurek, R. ; Sojka, P.: Software framework for topic modelling with large corpora.
  In: http://radimrehurek.com/gensim/lrec2010_final.pdf.
  
  Abstract: Large corpora are ubiquitous in today's world and memory quickly becomes the limiting factor in practical applications of the Vector Space Model (VSM). In this paper, we identify a gap in existing implementations of many of the popular algorithms, which is their scalability and ease of use. We describe a Natural Language Processing software framework which is based on the idea of document streaming, i.e. processing corpora document after document, in a memory independent fashion. Within this framework, we implement several popular algorithms for topical inference, including Latent Semantic Analysis and Latent Dirichlet Allocation, in a way that makes them completely independent of the training corpus size. Particular emphasis is placed on straightforward and intuitive framework design, so that modifications and extensions of the methods and/or their application by interested practitioners are effortless. We demonstrate the usefulness of our approach on a real-world scenario of computing document similarities within an existing digital library DML-CZ.
  
  Inhalt: Für die Software, vgl.: http://radimrehurek.com/gensim/index.html. Für eine Demo, vgl.: http://dml.cz/handle/10338.dmlcz/100785/SimilarArticles.
  
  Wissenschaftsfach: Mathematik
  Objekt: Latent Semantic Indexing
  Hilfsmittel: Gensim
  
  
• 10
  Kwakkel, J.H. ; Cunningham, S.W.: Managing polysemy and synonymy in science mapping using the mixtures of factor analyzers model.
  In: Journal of the American Society for Information Science and Technology. 60(2009) no.10, S.2064-2078.
  
  Abstract: A new method for mapping the semantic structure of science is described. We assume that different researchers, working on the same set of research problems, will use the same words for concepts central to their research problems. Therefore, different research fields and disciplines should be identifiable by different words and the pattern of co-occurring words. In natural language, however, there is quite some diversity because many words have multiple meaning. In addition, the same meaning can be expressed by using different words. We argue that traditional factor analytic and cluster analytic techniques are inadequate for mapping the semantic structure if such polysemous and synonymous words are present. Instead, an alternative model, the mixtures of factor analyzers (MFA) model, is utilized. This model extends the traditional factor analytic model by allowing multiple centroids of the dataset. We argue that this model is structurally better suited to map the semantic structure of science. The model is illustrated by a case study of the uncertainty literature sampled from data from the ISI Web of Science. The MFA model is applied with the goal of discovering multiple, potentially incommensurate, conceptualizations of uncertainty in the literature. In this way, the MFA model can help in creating understanding of the use of language in science, which can benefit multidisciplinary research and interdisciplinary understanding, and assist in the development of multidisciplinary taxonomies of science.
  
  Objekt: Latent Semantic Indexing
  
  
• 11
  Li, D. ; Kwong, C.-P. ; Lee, D.L.: Unified linear subspace approach to semantic analysis.
  In: Journal of the American Society for Information Science and Technology. 61(2010) no.1, S.175-189.
  
  Abstract: The Basic Vector Space Model (BVSM) is well known in information retrieval. Unfortunately, its retrieval effectiveness is limited because it is based on literal term matching. The Generalized Vector Space Model (GVSM) and Latent Semantic Indexing (LSI) are two prominent semantic retrieval methods, both of which assume there is some underlying latent semantic structure in a dataset that can be used to improve retrieval performance. However, while this structure may be derived from both the term space and the document space, GVSM exploits only the former and LSI the latter. In this article, the latent semantic structure of a dataset is examined from a dual perspective; namely, we consider the term space and the document space simultaneously. This new viewpoint has a natural connection to the notion of kernels. Specifically, a unified kernel function can be derived for a class of vector space models. The dual perspective provides a deeper understanding of the semantic space and makes transparent the geometrical meaning of the unified kernel function. New semantic analysis methods based on the unified kernel function are developed, which combine the advantages of LSI and GVSM. We also prove that the new methods are stable because although the selected rank of the truncated Singular Value Decomposition (SVD) is far from the optimum, the retrieval performance will not be degraded significantly. Experiments performed on standard test collections show that our methods are promising.
  
  Themenfeld: Semantisches Umfeld in Indexierung u. Retrieval
  Objekt: Latent Semantic Indexing ; Generalized Vector Space Model
  
  
• 12
  Martin, D.I. ; Berry, M.W.: Latent Semantic Indexing.
  In: Encyclopedia of library and information sciences. 3rd ed. Ed.: M.J. Bates. London : Taylor & Francis, 2009. S.3195-3204.
  
  Abstract: Latent Semantic Indexing (LSI) is a proven successful indexing and retrieval method. This method is based on an automated, mathematical technique known as singular value decomposition (SVD). Given a large information database, LSI uses SVD to create a "semantic space" of the document collection where both terms and documents are represented. It does this by producing a reduced dimensional vector space in which the underlying or "latent" semantic structure in the pattern of word usage of the document collection emerges. Similarities between terms, terms and documents, or documents in the document collection are then based on semantic content not on individual terms. This ability to extract meaning of terms and documents has given LSI success in many different applications.
  
  Inhalt: Digital unter: http://dx.doi.org/10.1081/E-ELIS3-120044505. Vgl.: http://www.tandfonline.com/doi/book/10.1081/E-ELIS3.
  
  Objekt: Latent semantic indexing
  
  
• 13
  Efron, M.: Query expansion and dimensionality reduction : Notions of optimality in Rocchio relevance feedback and latent semantic indexing.
  In: Information processing and management. 44(2008) no.1, S.163-180.
  
  Abstract: Rocchio relevance feedback and latent semantic indexing (LSI) are well-known extensions of the vector space model for information retrieval (IR). This paper analyzes the statistical relationship between these extensions. The analysis focuses on each method's basis in least-squares optimization. Noting that LSI and Rocchio relevance feedback both alter the vector space model in a way that is in some sense least-squares optimal, we ask: what is the relationship between LSI's and Rocchio's notions of optimality? What does this relationship imply for IR? Using an analytical approach, we argue that Rocchio relevance feedback is optimal if we understand retrieval as a simplified classification problem. On the other hand, LSI's motivation comes to the fore if we understand it as a biased regression technique, where projection onto a low-dimensional orthogonal subspace of the documents reduces model variance.
  
  Themenfeld: Retrievalalgorithmen
  Objekt: Rocchio-Algorithmus ; Latent semantic indexing
  
  
• 14
  Zhan, J. ; Loh, H.T.: Using latent semantic indexing to improve the accuracy of document clustering.
  In: Journal of information and knowledge management. 6(2007) no.3, S.181-188.
  
  Abstract: Document clustering is a significant research issue in information retrieval and text mining. Traditionally, most clustering methods were based on the vector space model which has a few limitations such as high dimensionality and weakness in handling synonymous and polysemous problems. Latent semantic indexing (LSI) is able to deal with such problems to some extent. Previous studies have shown that using LSI could reduce the time in clustering a large document set while having little effect on clustering accuracy. However, when conducting clustering upon a small document set, the accuracy is more concerned than efficiency. In this paper, we demonstrate that LSI can improve the clustering accuracy of a small document set and we also recommend the dimensions needed to achieve the best clustering performance.
  
  Objekt: Latent Semantic Indexing
  
  
• 15
  Rishel, T. ; Perkins, L.A. ; Yenduri, S. ; Zand, F.: Determining the context of text using augmented latent semantic indexing.
  In: Journal of the American Society for Information Science and Technology. 58(2007) no.14, S.2197-2204.
  
  Abstract: Latent semantic analysis has been used for several years to improve the performance of document library searches. We show that latent semantic analysis, augmented with a Part-of-Speech Tagger, may be an effective algorithm for classifying a textual document as well. Using Brille's Part-of-Speech Tagger, we truncate the singular value decomposition used in latent semantic analysis to reduce the size of the word-frequency matrix. This method is then tested on a toy problem, and has shown to increase search accuracy. We then relate these results to natural language processing and show that latent semantic analysis can be combined with context free grammars to infer semantic meaning from natural language. English is the natural language currently being used.
  
  Objekt: Latent semantic indexing
  
  
• 16
  Puzicha, J.: Informationen finden! : Intelligente Suchmaschinentechnologie & automatische Kategorisierung.
  Rheinbach : recommind, 2007. 15 S.
  
  Abstract: Wie in diesem Text erläutert wurde, ist die Effektivität von Such- und Klassifizierungssystemen durch folgendes bestimmt: 1) den Arbeitsauftrag, 2) die Genauigkeit des Systems, 3) den zu erreichenden Automatisierungsgrad, 4) die Einfachheit der Integration in bereits vorhandene Systeme. Diese Kriterien gehen davon aus, dass jedes System, unabhängig von der Technologie, in der Lage ist, Grundvoraussetzungen des Produkts in Bezug auf Funktionalität, Skalierbarkeit und Input-Methode zu erfüllen. Diese Produkteigenschaften sind in der Recommind Produktliteratur genauer erläutert. Von diesen Fähigkeiten ausgehend sollte die vorhergehende Diskussion jedoch einige klare Trends aufgezeigt haben. Es ist nicht überraschend, dass jüngere Entwicklungen im Maschine Learning und anderen Bereichen der Informatik einen theoretischen Ausgangspunkt für die Entwicklung von Suchmaschinen- und Klassifizierungstechnologie haben. Besonders jüngste Fortschritte bei den statistischen Methoden (PLSA) und anderen mathematischen Werkzeugen (SVMs) haben eine Ergebnisqualität auf Durchbruchsniveau erreicht. Dazu kommt noch die Flexibilität in der Anwendung durch Selbsttraining und Kategorienerkennen von PLSA-Systemen, wie auch eine neue Generation von vorher unerreichten Produktivitätsverbesserungen.
  
  Inhalt: Technical Whitepaper - Grundlagen der Informationsgewinnung
  
  Anmerkung: Vgl. auch: http://www.recommind.de/?id=mindserver_categorization.
  
  Themenfeld: Automatisches Klassifizieren
  Objekt: Latent Semantic Indexing
  
  
• 17
  Kontostathis, A. ; Pottenger, W.M.: ¬A framework for understanding Latent Semantic Indexing (LSI) performance.
  In: Information processing and management. 42(2006) no.1, S.56-73.
  
  Abstract: In this paper we present a theoretical model for understanding the performance of Latent Semantic Indexing (LSI) search and retrieval application. Many models for understanding LSI have been proposed. Ours is the first to study the values produced by LSI in the term by dimension vectors. The framework presented here is based on term co-occurrence data. We show a strong correlation between second-order term co-occurrence and the values produced by the Singular Value Decomposition (SVD) algorithm that forms the foundation for LSI. We also present a mathematical proof that the SVD algorithm encapsulates term co-occurrence information.
  
  Anmerkung: Beitrag innerhalb eines thematischen Schwerpunktes "Formal Methods for Information Retrieval"
  
  Objekt: Latent Semantic Indexing
  
  
• 18
  Bradford, R.B.: Relationship discovery in large text collections using Latent Semantic Indexing.
  In: Proceedings of the Fourth Workshop on Link Analysis, Counterterrorism, and Security, SIAM Data Mining Conference, Bethesda, MD, 20-22 April, 2006. [http://www.siam.org/meetings/sdm06/workproceed/Link%20Analysis/15.pdf].
  
  Abstract: This paper addresses the problem of information discovery in large collections of text. For users, one of the key problems in working with such collections is determining where to focus their attention. In selecting documents for examination, users must be able to formulate reasonably precise queries. Queries that are too broad will greatly reduce the efficiency of information discovery efforts by overwhelming the users with peripheral information. In order to formulate efficient queries, a mechanism is needed to automatically alert users regarding potentially interesting information contained within the collection. This paper presents the results of an experiment designed to test one approach to generation of such alerts. The technique of latent semantic indexing (LSI) is used to identify relationships among entities of interest. Entity extraction software is used to pre-process the text of the collection so that the LSI space contains representation vectors for named entities in addition to those for individual terms. In the LSI space, the cosine of the angle between the representation vectors for two entities captures important information regarding the degree of association of those two entities. For appropriate choices of entities, determining the entity pairs with the highest mutual cosine values yields valuable information regarding the contents of the text collection. The test database used for the experiment consists of 150,000 news articles. The proposed approach for alert generation is tested using a counterterrorism analysis example. The approach is shown to have significant potential for aiding users in rapidly focusing on information of potential importance in large text collections. The approach also has value in identifying possible use of aliases.
  
  Inhalt: Vgl. auch: http://www.contentanalyst.com/html/whoweare/whitepapers/whitepaper_why_lsi_latent_semantic_indexing_information_and_retrieval.html.
  
  Themenfeld: Semantisches Umfeld in Indexierung u. Retrieval
  Objekt: Latent Semantic Indexing
  
  
• 19
  Ding, C.H.Q.: ¬A probabilistic model for Latent Semantic Indexing.
  In: Journal of the American Society for Information Science and Technology. 56(2005) no.6, S.597-608.
  
  Abstract: Latent Semantic Indexing (LSI), when applied to semantic space built an text collections, improves information retrieval, information filtering, and word sense disambiguation. A new dual probability model based an the similarity concepts is introduced to provide deeper understanding of LSI. Semantic associations can be quantitatively characterized by their statistical significance, the likelihood. Semantic dimensions containing redundant and noisy information can be separated out and should be ignored because their negative contribution to the overall statistical significance. LSI is the optimal solution of the model. The peak in the likelihood curve indicates the existence of an intrinsic semantic dimension. The importance of LSI dimensions follows the Zipf-distribution, indicating that LSI dimensions represent latent concepts. Document frequency of words follows the Zipf distribution, and the number of distinct words follows log-normal distribution. Experiments an five standard document collections confirm and illustrate the analysis.
  
  Themenfeld: Retrievalstudien
  Objekt: Latent Semantic Indexing
  
  
• 20
  Efron, M.: Eigenvalue-based model selection during Latent Semantic Indexing.
  In: Journal of the American Society for Information Science and Technology. 56(2005) no.9, S.969-988.
  
  Abstract: In this study amended parallel analysis (APA), a novel method for model selection in unsupervised learning problems such as information retrieval (IR), is described. At issue is the selection of k, the number of dimensions retained under latent semantic indexing (LSI). Amended parallel analysis is an elaboration of Horn's parallel analysis, which advocates retaining eigenvalues larger than those that we would expect under term independence. Amended parallel analysis operates by deriving confidence intervals an these "null" eigenvalues. The technique amounts to a series of nonparametric hypothesis tests an the correlation matrix eigenvalues. In the study, APA is tested along with four established dimensionality estimators an six Standard IR test collections. These estimates are evaluated with regard to two IR performance metrics. Additionally, results from simulated data are reported. In both rounds of experimentation APA performs weIl, predicting the best values of k an 3 of 12 observations, with good predictions an several others, and never offering the worst estimate of optimal dimensionality.
  
  Objekt: Latent Semantic Indexing