How do global warming and agriculture influence each other? It is possible to answer the question by searching knowledge about the relationship between global warming and agriculture. As exemplified by this question, strong demands exist for searching relationships between objects. Mining knowledge about relationships on Wikipedia has been studied. However, it is desired to search more diverse knowledge about relationships on the Web. By utilizing the objects constituting relationships mined from Wikipedia, we propose a new method to search images with surrounding text that include knowledge about relationships on the Web. Experimental results show that our method is effective and applicable in searching knowledge about relationships. We also construct a relationship search system named “Enishi” based on the proposed new method. Enishi supplies a wealth of diverse knowledge including images with surrounding text to help users to understand relationships deeply, by complementarily utilizing knowledge from Wikipedia and the Web.

We present an algorithm for simultaneously recognizing and localizing planar textured objects in an image. The algorithm can scale efficiently with respect to a large number of objects added into the database. In contrast to the current state-of-the-art on large scale image search, our algorithm can accurately work with query images consisting of several specific objects and/or multiple instances of the same object. Our proposed algorithm consists of two major steps. The first step is to generate a set of hypotheses that provides information about the identities and the locations of objects in the image. To serve this purpose, we extend Bag-Of-Visual-Word (BOVW) image retrieval by incorporating a re-ranking scheme based on the Hough voting technique. Subsequently, in the second step, we propose a geometric verification algorithm based on A Contrario decision framework to draw out the final detection results from the generated hypotheses. We demonstrate the performance of the algorithm on the scenario of recognizing CD covers with a database consisting of more than ten thousand images of different CD covers. Our algorithm yield to the detection results of more than 90% precision and recall within a few seconds of processing time per image.

The topic of retrieving videos containing a desired person from a dataset just using the content of faces without any help of textual information has many interesting applications like video surveillance, social network, video mining, etc. However, traditional face matching against a huge number of detected faces leads to an unacceptable response time and may also reduce the accuracy due to the large variations in facial expressions, poses, lighting, etc. Therefore, in this paper we propose a novel method to generate discriminative “signatures” for efficiently retrieving the videos containing the same person with a query. In this research, the signature is defined as a compact, discriminative and reduced dimensionality representation, which is generated from a set of high-dimensional feature vectors of an individual. The desired videos are retrieved based on the similarities between the signature of the query and those of individuals in the database. In particular, we make the following contributions. Firstly, we give an algorithm of two directional linear discriminant analysis with maximum correntropy criterion (2DLDA-MCC) as an extension to our recently proposed maximum correntropy criterion based linear discriminant analysis (LDA-MCC). Both algorithms are robust to outliers and noise. Secondly, we present an approach for transferring a set of exemplars to a fixed-length signature using LDA-MCC and 2DLDA-MCC, resulting in two kinds of signatures that are called 1D signature and 2D signature. Finally, a novel video retrieval scheme is given based on the signatures, which has low storage requirement and can achieve a fast search. Evaluations on a large dataset of videos show reliable measurement of similarities by using the proposed signatures to represent the identities generated from videos. Experimental results also demonstrate that the proposed video retrieval scheme has the potential to substantially reduce the response time and slightly increase the mean average precision of retrieval.

Comparable Corpora are valuable resources for many NLP applications, and extensive research has been done on information mining based on comparable corpora in recent years. While there are not enough large-scale available public comparable corpora at present, this paper presents a bi-directional CLIR-based method for creating comparable corpora from two independent news collections in different languages. The original Chinese document collections and English documents collections are crawled from XinHuaNet respectively and formatted in a consistent manner. For each document from the two collections, the best query keywords are extracted to represent the essential content of the document, and then the keywords are translated into the language of the other collection. The translated queries are run against the collection in the same language to pick up the candidate documents in the other language and candidates are aligned based on their publication dates and the similarity scores. Results show that our approach significantly outperforms previous approaches to the construction of Chinese-English comparable corpora.

Music-similarity computation is an essential building block for the browsing, retrieval, and indexing of digital music archives. This paper proposes a music similarity function based on the centroid model, which divides the feature space into non-overlapping clusters for the efficient computation of the timber distance of two songs. We place particular emphasis on the centroid deviation as a feature for music-similarity computation. Experiments show that the centroid-model representation of the auditory features is promising for music-similarity computation.

This Letter presents a new feature named structured local binary Kirsch pattern (SLBKP) for image retrieval. Each input color image is decomposed into Y, Cb and Cr components. For each component image, eight 33 Kirsch direction templates are first performed pixel by pixel, and thus each pixel is characterized by an 8-dimenional edge-strength vector. Then a binary operation is performed on each edge-strength vector to obtain its integer-valued SLBKP. Finally, three SLBKP histograms are concatenated together as the final feature of each input colour image. Experimental results show that, compared with the existing structured local binary Haar pattern (SLBHP)-based feature, the proposed feature can greatly improve retrieval performance.

We propose a retrieval method using scale invariant visual phrases (SIVPs). Our method encodes spatial information into the SIVPs which capture translation, rotation and scale invariance, and employs the SIVPs to determine the spatial correspondences between query image and database image. To compute the spatial correspondences efficiently, the SIVPs are introduced into the inverted index, and SIVP verification is investigated to refine the candidate images returned from inverted index. Experimental results demonstrate that our method improves the retrieval accuracy while increasing the retrieval efficiency.

There has recently been much research on content-based image retrieval (CBIR) that uses image features including color, shape, and texture. In CBIR, feature extraction is important because the retrieval result depends on the image feature. Query-by-sketch image retrieval is one of CBIR and query-by-sketch image retrieval is efficient because users simply have to draw a sketch to retrieve the desired images. In this type of retrieval, selecting the optimum feature extraction method is important because the retrieval result depends on the image feature. We have developed a query-by-sketch image retrieval method that uses an edge relation histogram (ERH) as a global and local feature intended for binary line images. This histogram is based on the patterns of distribution of other line pixels centered on each line pixel that have been obtained by global and local processing. ERH, which is a shift- and scale-invariant feature, focuses on the relation among the edge pixels. It is fairly simple to describe rotation- and symmetry-invariant features, and query-by-sketch image retrieval using ERH makes it possible to perform retrievals that are not affected by position, size, rotation, or mirroring. We applied the proposed method to 20,000 images in the Corel Photo Gallery. Experimental results showed that it was an effective means of retrieving images.

Shape is one of the primary low-level image features in content-based image retrieval. In this paper we propose a new shape description method that consists of a rotationally invariant angular radial transform descriptor (IARTD). The IARTD is a feature vector that combines the magnitude and aligned phases of the angular radial transform (ART) coefficients. A phase correction scheme is employed to produce the aligned phase so that the IARTD is invariant to rotation. The distance between two IARTDs is defined by combining differences in the magnitudes and aligned phases. In an experiment using the MPEG-7 shape dataset, the proposed method outperforms existing methods; the average BEP of the proposed method is 57.69%, while the average BEPs of the invariant Zernike moments descriptor and the traditional ART are 41.64% and 36.51%, respectively.

Polar and Spherical Fourier analysis can be used to extract rotation invariant features for image retrieval and pattern recognition tasks. They are demonstrated to show superiorities comparing with other methods on describing rotation invariant features of two and three dimensional images. Based on mathematical properties of trigonometric functions and associated Legendre polynomials, fast algorithms are proposed for multimedia applications like real time systems and large multimedia databases in order to increase the computation speed. The symmetric points are computed simultaneously. Inspired by relative prime number theory, systematic analysis are given in this paper. Novel algorithm is deduced that provide even faster speed. Proposed method are 9–15% faster than previous work. The experimental results on two and three dimensional images are given to illustrate the effectiveness of the proposed method. Multimedia signal processing applications that need real time polar and spherical Fourier analysis can be benefit from this work.

We are interested in retrieving video shots or videos containing particular people from a video dataset. Owing to the large variations in pose, illumination conditions, occlusions, hairstyles and facial expressions, face tracks have recently been researched in the fields of face recognition, face retrieval and name labeling from videos. However, when the number of face tracks is very large, conventional methods, which match all or some pairs of faces in face tracks, will not be effective. Therefore, in this paper, an efficient method for finding a given person from a video dataset is presented. In our study, in according to performing research on face tracks in a single video, we also consider how to organize all the faces in videos in a dataset and how to improve the search quality in the query process. Different videos may include the same person; thus, the management of individuals in different videos will be useful for their retrieval. The proposed method includes the following three points. (i) Face tracks of the same person appearing for a period in each video are first connected on the basis of scene information with a time constriction, then all the people in one video are organized by a proposed hierarchical clustering method. (ii) After obtaining the organizational structure of all the people in one video, the people are organized into an upper layer by affinity propagation. (iii) Finally, in the process of querying, a remeasuring method based on the index structure of videos is performed to improve the retrieval accuracy. We also build a video dataset that contains six types of videos: films, TV shows, educational videos, interviews, press conferences and domestic activities. The formation of face tracks in the six types of videos is first researched, then experiments are performed on this video dataset containing more than 1 million faces and 218,786 face tracks. The results show that the proposed approach has high search quality and a short search time.

This paper describes the application of two attractive categories of topic modeling techniques to the problem of spoken document retrieval (SDR), viz. document topic model (DTM) and word topic model (WTM). Apart from using the conventional unsupervised training strategy, we explore a supervised training strategy for estimating these topic models, imagining a scenario that user query logs along with click-through information of relevant documents can be utilized to build an SDR system. This attempt has the potential to associate relevant documents with queries even if they do not share any of the query words, thereby improving on retrieval quality over the baseline system. Likewise, we also study a novel use of pseudo-supervised training to associate relevant documents with queries through a pseudo-feedback procedure. Moreover, in order to lessen SDR performance degradation caused by imperfect speech recognition, we investigate leveraging different levels of index features for topic modeling, including words, syllable-level units, and their combination. We provide a series of experiments conducted on the TDT (TDT-2 and TDT-3) Chinese SDR collections. The empirical results show that the methods deduced from our proposed modeling framework are very effective when compared with a few existing retrieval approaches.

In this paper, we propose a duplicate document detection model recognizing both partial duplicates and near duplicates. The proposed model can detect partial duplicates as well as exact duplicates by splitting a large document into many small sentence fingerprints. Furthermore, the proposed model can detect even near duplicates, the result of trivial revisions, by filtering the common words and reordering the word sequence.

Feature location is to identify source code that implements a given feature. It is essential for software maintenance and evolution. A large amount of research, including static analysis, dynamic analysis and the hybrid approaches, has been done on the feature location problems. The existing approaches either need plenty of scenarios or rely on domain experts heavily. This paper proposes a new approach to locate functional feature in source code by combining the change impact analysis and information retrieval. In this approach, the source code is instrumented and executed using a single scenario to obtain the execution trace. The execution trace is extended according to the control flow to cover all the potentially relevant classes. The classes are ranked by trace-based impact analysis and information retrieval. The ranking analysis takes advantages of the semantics and structural characteristics of source code. The identified results are of higher precision than the individual approaches. Finally, two open source cases have been studied and the efficiency of the proposed approach is verified.

A two-dimensional shape is generally represented with line drawings or object contours in a digital image. Shapes can be divided into two types, namely ordered and unordered shapes. An ordered shape is an ordered set of points, while an unordered shape is an unordered set. As a result, each type typically uses different attributes to define the local descriptors involved in representing the local distributions of points sampled from the shape. Throughout this paper, we focus on unordered shapes. Since most local descriptors of unordered shapes are not scale-invariant, we usually make the shapes in an image data set the same size through scale normalization, before applying shape matching procedures. Shapes obtained through scale normalization are suitable for such descriptors if the original whole shapes are similar. However, they are not suitable if parts of each original shape are drawn using different scales. Thus, in this paper, we present a scale-invariant descriptor constructed by von Mises distributions to deal with such shapes. Since this descriptor has the merits of being both scale-invariant and a probability distribution, it does not require scale normalization and can employ an arbitrary measure of probability distributions in matching shape points. In experiments on shape matching and retrieval, we show the effectiveness of our descriptor, compared to several conventional descriptors.

Kernel biased discriminant analysis (KBDA), as a subspace learning algorithm, has been an attractive approach for the relevance feedback in content-based image retrieval. Its performance, however, still suffers from the “small sample learning” problem and “kernel learning” problem. Aiming to solve these problems, in this paper, we present a new semi-supervised scheme of KBDA (S-KBDA), in which the projection learning and the “kernel learning” are interweaved into a constrained optimization framework. Specifically, S-KBDA learns a subspace that preserves both the biased discriminant structure among the labeled samples, and the geometric structure among all training samples. In kernel optimization, we directly optimize the kernel matrix, rather than a kernel function, which makes the kernel learning more flexible and appropriate for the retrieval task. To solve the constrained optimization problem, a fast algorithm based on gradient ascent is developed. The image retrieval experiments are given to show the effectiveness of the S-KBDA scheme in comparison with the original KBDA, and the other two state-of-the-art algorithms.

Learning to rank refers to machine learning techniques for training the model in a ranking task. Learning to rank is useful for many applications in Information Retrieval, Natural Language Processing, and Data Mining. Intensive studies have been conducted on the problem and significant progress has been made [1],[2]. This short paper gives an introduction to learning to rank, and it specifically explains the fundamental problems, existing approaches, and future work of learning to rank. Several learning to rank methods using SVM techniques are described in details.

Networks have gotten bigger recently, and users have a more difficult time finding the information that they want. The use of mobile agents to help users effectively retrieve information has garnered a lot of attention. In this paper, we propose an agent control method for time constrained information retrieval. We pay attention to the highest past score gained by the agents and control the agents with the expectation of achieving better scores. Using computer simulations, we confirmed that our control method gave the best improvement over the whole network while reducing the overall variance. From these results, we can say that our control method improves the quality of information retrieved by the agent.

This paper introduces a future and past search engine, ChronoSeeker, which can help users to develop long-term strategies for their organizations. To provide on-demand searches, we tackled two technical issues: (1) organizing efficient event searches and (2) filtering out noises from search results. Our system employed query expansion with typical expressions related to event information such as year expressions, temporal modifiers, and context terms for efficient event searches. We utilized a machine-learning technique of filtering noise to classify candidates into information or non-event information, using heuristic features and lexical patterns derived from a text-mining approach. Our experiment revealed that filtering achieved an 85% F-measure, and that query expansion could collect dozens more events than those without expansion.

Based on sliding-window rule application and extraction filtering, we present a framework for extracting multi-slot frames describing chemical reactions from Thai free text with unknown target-phrase boundaries. A supervised rule learning algorithm is employed for automatic construction of pattern-based extraction rules from hand-tagged training phrases. A filtering method is devised for removal of incorrect extraction results based on features observed from text portions appearing between adjacent slot fillers in source documents. Extracted reaction frames are represented as concept expressions in description logics and are used as metadata for document indexing. A document knowledge base supporting semantics-based information retrieval is constructed by integrating document metadata with domain-specific ontologies.