Using context to build semantics

In recent years, latent semantic analysis (LSA) has reached a level of maturity at which its presence is ubiquitous in technology as well as in simulation of cognitive processes. In spite of this, in recent years there has been a trend of subjecting LSA to some criticisms, usually because it is compared to other models in very specific tasks and conditions and sometimes without having good knowledge of what the semantic representation of LSA means, and without exploiting all the possibilities of which LSA is capable other than the cosine. This paper provides a critical review to clarify some of the misunderstandings regarding LSA and other space models. The historical stability of the predecessors of LSA, the representational structure of word meaning and the multiple topologies that could arise from a semantic space, the computation of similarity, the myth that LSA dimensions have no meaning, the computational and algorithm plausibility to account for meaning acquisition in LSA (in contrast to others models based on online mechanisms), the possibilities of spatial models to substantiate recent proposals, and, in general, the characteristics of classic vector models and their ease and flexibility to simulate some cognitive phenomena will be reviewed. The review highlights the similarity between LSA and other techniques and proposes using long LSA experiences in other models, especially in predicting models such as word2vec. In sum, it emphasizes the lessons that can be learned from comparing LSA-based models to other models, rather than making statements about "the best."

Nowadays, searching for a topic on the Internet can be a frustrating experience because of all the excessive information. Thus, a strategy for automatically classifying the results can improve user experience and work efficiency. Latent Semantic Indexing (LSI) algorithm is used to classify documents by meaning due to its effectiveness. However, there is a problem with the implementation of this algorithm. LSI is computationally intensive because the cost is directly related to the number of documents. In particular, the Singular Value Decomposition (SVD) that is mainly used in LSI is unscalable in terms of both memory and computation time. One possible solution is to use more powerful computational resources, such as multiple computing nodes. In this paper, a novel distributed architecture for the LSI algorithm is proposed. It is based on the use of microservices in a Google Cloud environment. We evaluated the performances of the proposed Cloud-based LSI, and comparison is made with standalone LSI. The results show the benefits of using distributed systems based on runtime, concurrency, and processing.

Job matching is a hiring process that involves a thorough understanding of the context and meaning of words in different languages. The updated and expanded latent semantic indexing (LSI) Framework seeks to improve the precision and relevance of job matching analysis of word meanings in multi-languages. Because they only compare related terms, conventional LSIs are often insufficient to address the complexity of context in job matching. Extending the LSI approach can improve the vector representation of words and help you understand the context and semantic relationships in the text. Improved LSI analyzes context more precisely by using word vector representation. Improved LSI focuses on understanding semantic relationships between words in many languages to produce more accurate and relevant job matches. This paper describes the steps involved in improving LSI, such as data collection, pre-processing, linguistic feature extraction, LSI model training, and evaluation of matching results. The results show that the examined classification model has much better performance in terms of word classification. Conventional LSI has an average prediction value of 79%, once the enhanced LSI can accurately predict about 84% of the entire word, it has a reasonable capacity to recognize the actual words in a natural context.

We present an application of a novel methodology called Text Influenced Molecular Indexing (TIMI) to mine the information in the scientific literature. TIMI is an extension of two existing methodologies: (1) Latent Semantic Structure Indexing (LaSSI), a method for calculating chemical similarity using two-dimensional topological descriptors, and (2) Latent Semantic Indexing (LSI), a method for generating correlations between textual terms. The singular value decomposition (SVD) of a feature/object matrix is the fundamental mathematical operation underlying LSI, LaSSI, and TIMI and is used in the identification of associations between textual and chemical descriptors. We present the results of our studies with a database containing 11,571 PubMed/MEDLINE abstracts which show the advantages of merging textual and chemical descriptors over using either text or chemistry alone. Our work demonstrates that searching text-only databases limits retrieved documents to those that explicitly mention compounds by name in the text. Similarly, searching chemistry-only databases can only retrieve those documents that have chemical structures in them. TIMI, however, enables search and retrieval of documents with textual, chemical, and/or text- and chemistry-based queries. Thus, the TIMI system offers a powerful new approach to uncovering the contextual scientific knowledge sought by the medical research community.

Text representation, which is a fundamental and necessary process for text-based intelligent information processing, includes the tasks of determining the index terms for documents and producing the numeric vectors corresponding to the documents. In this paper, multi-word, which is regarded as containing more contextual semantics than individual word and possessing the favorable statistical characteristics, is proposed as an alternative index terms in vector space model for text representation with theoretical support. We investigate the traditional indexing methods as TF*IDF (term frequency inverse document frequency) and LSI (latent semantic indexing) for comparative study. The performances of TF*IDF, LSI and multi-word are examined on the tasks of text classification, which includes information retrieval (IR) and text categorization (TC), in Chinese and English document collection respectively. We also attempt to tune the rescaling factor of LSI and observe its effectiveness in text classification. The experimental results demonstrate that TF*IDF and multi-word are comparable when they are used for IR and TC and LSI is the poorest one of them. Moreover, the rescaling factor of LSI has an insignificant influence on its effectiveness on text classification for both Chinese and English text classification.

The exponential growth of data demands scalable infrastructures capable of indexing and searching rich content such as text, music, and images. A promising direction is to combine information re-trieval with peer-to-peer technology for scalability, fault-tolerance, and low administration cost. One pioneering work along this di-rection is pSearch [32, 33]. pSearch places documents onto a peer-to- peer overlay network according to semantic vectors produced using Latent Semantic Indexing (LSI). The search cost for a query is reduced since documents related to the query are likely to be co-located on a small number of nodes. Unfortunately, because of its reliance on LSI, pSearch also inherits the limitations of LSI. (1) When the corpus is large and heterogeneous, LSI's retrieval quality is inferior to methods such as Okapi. (2) The Singular Value Decomposition (SVD) used in LSI is unscalable in terms of both memory consumption and computation time.This paper addresses the above limitations of LSI and makes the following contributions. (1) To reduce the cost of SVD, we reduce the size of its input matrix through document clustering and term selection. Our method retains the retrieval quality of LSI but is several orders of magnitude more efficient. (2) Through extensive experimentation, we found that proper normalization of semantic vectors for terms and documents improves recall by 76%. (3) To further improve retrieval quality, we use low-dimensional subvectors of semantic vectors to cluster documents in the overlay and then use Okapi to guide the search and document selection.

ldquoCurse of dimensionalityrdquo is a common problem in the area of information retrieval. It was verified that points in a vector space are projected to a random subspace of suitably high dimension, and then the distances between the points are approximately preserved. Although such a random projection can be used to reduce the dimension of the document space, it does not bring together semantically related documents. Latent Semantic Indexing (LSI) projects documents to lower dimensional LSI space from higher dimensional term space with singular-value decomposition (SVD) for the purpose of reducing the dimensions of the document space and bringing together semantically related documents. But the computation time of SVD is a bottleneck because of the higher dimensions of documents. In this paper, a novel method of dimension reduction for improving LSI is provided. A term-to-concept projection matrix based on domain ontology was created in this method. This way documents were projected to lower dimensional concept space by the projection matrix. LSI pre-computation was performed not on the original term by document matrix, but on the lower dimensional concept by document matrix at great computational savings. Experiments indicate that this method improves the efficiency of LSI. And the similarity judgment between documents is not disturbed.

Singular value decomposition is an important method to factorize a matrix, which is useful for many applications, such as principal component analysis in statistics and latent semantic analysis in natural language text processing. In order to diminish the computational complexity, incremental update algorithm is developed, especially for analyzing relationships between text documents. However, it is at the cost of a reduction in precision. In this paper, we study the effect of incremental update on latent semantic analysis; and, by experiments, to investigate the trend downward of computing precision and then try to find the balance point between re-computation and incremental update. The results indicate that when increased documents are less than original documents, incremental update method has an approximate precision to re-computing method, and then its recall ratio will obviously decline. It could be helpful for deciding when the matrix factorization must be recomputed.

A framework for understanding Latent Semantic Indexing (LSI) performance

High-dimensional semantic space accounts of priming

In this chapter, we introduce latent semantic analysis (LSA), which uses singular value decomposition (SVD) to reduce the dimensionality of the document-term representation. This method reduces the large matrix to an approximation that is made up of fewer latent dimensions that can be interpreted by the analyst. Two important concepts in LSA, cosine similarity and queries, are explained. Finally, we discuss decision-making in LSA.

Internet is becoming an integral part of everyday life and email has become a powerful tool to exchange ideas and information. Through email users connect socially and commercially. But mass mailing i.e. Spam has become one of the biggest worldwide problem. So there is requirement to develop a spam filter which can detect the spam and can stop the mass mailing. In this paper, Latent Semantic Indexing (LSI) approach is used for spam filtering. Latent semantic indexing (LSI) is an indexing and retrieval method which uses singular value decomposition (SVD) to find the relationships between the terms and unstructured collection of message. LSI is based on the principle that terms that are used in the same contexts tend to have similar meanings. Apart from this spam are obtained more or less abundant replications in a significant numbers. The detection of these duplicates is important because it allows to lighten email box and will improve the efficiency of spam filter. The Duplicate Message Detection Techniques (DMDT) has been used to improve the efficiency and effectiveness of spam filter.

Latent Semantic Analysis (LSA) is a method that allows us to automatically index and retrieve information from a set of objects by reducing the term-by-document matrix using the Singular Value Decomposition (SVD) technique. However, LSA has a high computational cost for analyzing large amounts of information. The goals of this work are (i) to improve the execution time of semantic space construction, dimensionality reduction, and information retrieval stages of LSA based on heterogeneous systems and (ii) to evaluate the accuracy and recall of the information retrieval stage. We present a heterogeneous Latent Semantic Analysis (hLSA) system, which has been developed using General-Purpose computing on Graphics Processing Units (GPGPUs) architecture, which can solve large numeric problems faster through the thousands of concurrent threads on multiple CUDA cores of GPUs and multi-CPU architecture, which can solve large text problems faster through a multiprocessing environment. We execute the hLSA system with documents from the PubMed Central (PMC) database. The results of the experiments show that the acceleration reached by the hLSA system for large matrices with one hundred and fifty thousand million values is around eight times faster than the standard LSA version with an accuracy of 88% and a recall of 100%.

Document categorization is the process of classifying documents from many mixed documents automatically, and the main problem is how to express document content in vector space completely. This paper proposes a new model named Latent Semantic Analysis (LSA) + word2vec to categorize documents. This is the first attempt of combining word2vec with LSA at document categorization and it can map document to vector space under the premise of keeping document contents fully. At first, we create a term by document matrix and the element of which is decided by Term Frequency-Inverse Document Frequency (TF-IDF) weighting and word vector trained by word2vec. This matrix is a 3-dimensional matrix and it can describe the meaning of every word and the content of every document exactly. Secondly, Singular Value Decomposition (SVD) is executed on the matrix and lower computational complexity is gained from this. The model is named LSA + word2vec. Then, document vector gained from the new model are put into Convolutional Neural Network (CNN) to train. CNN is an efficient deep learning algorithm, which improves the accuracy of classification greatly. We evaluate the performance based on the 20newsgroups corpus. The results show that our new model achieves better effects on document categorization tasks, and the accuracy made about 15% improvement than traditional methods, such as LSA and Vector Space Model (VSM).

The standard application of Latent Semantic Indexing (LSI), a well-known technique for information retrieval, requires the computation of a partial Singular Value Decomposition (SVD) of the term-document matrix. This computation becomes infeasible for large document collections, since it is very demanding both in terms of arithmetic operations and in memory requirements. This paper discusses two divide and conquer strategies, with the goal of alleviating these difficulties. Both strategies divide the document collection in subsets, perform relevance analysis on each subset, and conquer the analysis results to form the query response. Since each sub-problem resulting from the recursive division process has a smaller size, the processing of large scale document collections requires much fewer resources. In addition, the computation is highly parallel and can be easily adapted to a parallel computing environment. To reduce the computational cost, we perform the analysis on the subsets by using the Lanczos vectors instead of singular vectors as in the standard LSI method. This technique is far more efficient than the computation of the truncated SVD, while its accuracy is comparable. Experimental results confirm that the proposed divide and conquer strategies are effective for information retrieval problems.