Search

The rapid development of the Internet leads to the rapid growth of network information, we call it information explosion. The Internet is full of information, and it is difficult for users to find this information and useful knowledge of the ocean. The Web has become the world's largest information repository, and there is an urgent need for efficient access to the valuable knowledge of vast amounts of web information. The purpose of this paper is to study the process and algorithm analysis of text mining system based on artificial intelligence. This paper presents an algorithm of document feature acquisition based on genetic algorithm. Selecting suitable features is an important task in specific text classification and information retrieval. Finding appropriate feature vectors to represent the text will undoubtedly help with subsequent sorting and grouping. Based on the genetic algorithm of variable length chromosome, this paper improves the crossover, mutation and selection operations, and proposes an algorithm to obtain text feature vectors. This method has a wide range of applications and good results. [ABSTRACT FROM AUTHOR]

In the era of rapid paper publications in various venues, automatic citation recommendations would be highly useful to researchers when they write papers. Local citation recommendation aims to recommend possible papers to cite given local citation contexts. Previous work mainly computes the similarity score between citation contexts and cited papers on a one-to-one basis, which is quite time-consuming. We train a pair of neural network encoders that map citation contexts and all possible cited papers to the same vector space, respectively. After that, we index the positions of all cited papers in the vector space. This makes our process for searching recommended papers considerably faster. On the other hand, existing methods tend to recommend papers that are highly similar to each other, which makes recommendations lack diversity. Therefore, we extend our algorithm to perform multi-topic recommendations. We generate multi-topic training examples based on the index we mentioned earlier. Furthermore, we specially design a multi-group contrastive learning method to train our model so that it can distinguish different topics. Empirical experiments show that our model outperforms previous methods by a wide margin. Our model is also light weighted and has been deployed online so that researchers can use it to obtain recommended citations for their own paper in real-time. • Proposed FLCR algorithm with sentence-transformer & k-d tree. • Introduced multi-topic citation recommendation for diverse contexts. • Developed large-scale dataset with 1.7 million citation contexts for evaluation. • Demonstrated significant performance improvement over previous methods. • Deployed demo for real-time citation recommendations. [ABSTRACT FROM AUTHOR]

Abstract: This paper studies the problem of expertise matching with various constraints. Expertise matching, which aims to find the alignment between experts and queries, is a common problem in many applications such as conference paper–reviewer assignment, product–reviewer alignment, and product-endorser matching. Most existing methods formalize this problem as an information-retrieval problem and focus on finding a set of experts for each query independently. However, in real-world systems, various constraints are often needed to be considered. For example, in order to review a paper, it is desirable that there is at least one senior reviewer to guide the reviewing process. An important question is: “Can we design a framework to efficiently find the optimal solution for expertise matching under various constraints?” This paper explores such an approach by formulating the expertise matching problem in a constraint-based optimization framework. In the proposed framework, the problem of expertise matching is linked to a convex cost flow problem, which guarantees an optimal solution under various constraints. We also present an online matching algorithm to support incorporating user feedbacks in real time. The proposed approach has been evaluated on two different genres of expertise matching problems, namely conference paper–reviewer assignment and teacher–course assignment. Experimental results validate the effectiveness of the proposed approach. Based on the proposed method, we have also developed an online system for paper–reviewer suggestions, which has been used for paper–reviewer assignment in a top conference and feedbacks from the conference organizers are very positive. [Copyright &y& Elsevier]

Clustering is a traditional data mining problem that has attracted researchers from different disciplines because its solution can be applied to many useful problems in our daily life. Since the era of big data is coming, how to "reduce the computing time" of an "effective clustering algorithm" has been a promising research issue in recent years. Thus, this paper presents an effective clustering algorithm, by using the so-called searched information to determine later search directions, and then has it implemented on Spark to accelerate its response time for analyzing large-scale datasets. Simulation results show that the proposed algorithm provides a better result than the other clustering algorithms compared in this paper because it is less sensitive to the initial solutions. The simulation results further show that cloud computing platform is capable of enhancing the performance of the proposed algorithm. • This study presents a high-performance clustering algorithm based on searched experiences. • The results show that it is able to find a better result than traditional clustering algorithms. • We apply it to Spark to show its possibility to accelerate the response time on a cloud system. [ABSTRACT FROM AUTHOR]

Layout analysis is essential step in information extraction from scanned document images. In this paper we propose an algorithm for aligning layouts generated with different OCR engines. The main requirement is to always generate the same layout for the given document image regardless of OCR engine used for image processing. In that way information extraction from scanned documents, that is heavily dependent on fields positions in the document, does not depend on specific OCR engine. In other words, it is sufficient to maintain universal extractor knowledge and not necessary to train extractor explicitly with samples processed on specific OCR engine. The proposed algorithm can handle administrative documents with complex layout. [ABSTRACT FROM AUTHOR]

Query reformulation techniques are essential for information retrieval systems. These techniques eliminate the bad queries (short and ambiguous queries) and minimize users search time. There are two kinds of techniques in literature; the first uses the query logs with the click-through data to identify relevant queries and propose them to the user. The second group uses relevant terms extracted from different sources such as Wikipedia, Wordnet, and pseudo-relevant documents to expand the initial query and increase the likelihood with relevant documents. Both groups generate false queries because they do not consider the user interests and due to the assumption that the user clicks only on relevant results (first group) and that the top-k retrieved documents (top 5 to 15) are relevant (second group).In this paper, we propose a novel approach to reformulate the user queries using his profile (which contains the user interests). The approach retrieves documents related to the query from four different data sources (≈ 1000 doc). Then, it extracts the topics of these documents using the Lingo clustering algorithm (cluster the document with the same topic) and Text razor API (extract the potential topics of each cluster). Finally, it generates queries based on these topics and the user profile. The results show that the proposed approach outperforms the existing solution in P@10, P@20, and MAP. [ABSTRACT FROM AUTHOR]

Although in the last decade several fact-checking organizations have emerged to verify misinformation, fake news has continued to proliferate, especially through social media platforms. Even though adopting improved detection strategies is of utmost importance, the fact-checking process could be optimized by verifying whether a claim has been previously fact-checked. Despite some ad-hoc information retrieval approaches having been recently proposed, the utility of modern (neural) retrieval systems have not been investigated yet. In this paper, we consider the standard two-phases retriever-reranker architecture and benchmark different state-of-the-art techniques from the information retrieval and Q&A literature. We design several experiments on a real-world Twitter dataset to analyze the efficiency and the effectiveness of the benchmark approaches. Our results show that combining standard and neural approaches is the most promising research direction to improve retrievers performance and that complex (neural) rerankers might still be efficient in practice since there is no need to process a high number of documents to improve ranking performance. • We address the problem of detecting previously fact-checked information. • Considering the information retrieval task, we define a multi-stage retriever-reranker framework. • We benchmark the state-of-the-art retriever and re-ranker architectures. • We discuss the system's operational settings unveiling its strengths and weaknesses. [ABSTRACT FROM AUTHOR]

This work presents a novel automated approach to construct topic knowledge maps with knowledge structures, followed by its application to an internationally renowned journal. Knowledge structures are diagrams showing the important components of knowledge in study. Knowledge maps identify the locations of objects and illustrate the relationship among objects. In our study, the important components derived from knowledge structures are used as objects to be spotted in a topic knowledge map. The purpose of our knowledge structures is to find out the major topics serving as subjects of article collections as well as related methods employed in the published papers. The purpose of topic knowledge maps is to transform high-dimensional objects (topic, paper, and cited frequency) into a 2-dimensional space to help understand complicated relatedness among high-dimensional objects, such as the related degree between an article and a topic. First, we adopt independent chi-square test to examine the independence of topics and apply genetic algorithm to choose topics selection with best fitness value to construct knowledge structures. Additionally, high-dimensional relationships among objects are transformed into a 2-dimensional space using the multi-dimension scaling method. The optimal transformation coordinate matrix is also determined by using a genetic algorithm to preserve the original relations among objects and construct appropriate topic knowledge maps. [ABSTRACT FROM AUTHOR]

Abstract: Case-Based reasoning (CBR) applications are increasingly used for problems involving structured data, especially in process-oriented domains. Available CBR frameworks rely primarily on flat case representations, leaving developers of process-oriented CBR systems without general-purpose tools aimed at their needs. This paper discusses needs to support structure-based retrieval for CBR and how they have been addressed in the development of the Structure Access Interface (SAI), a toolkit for representation and retrieval of structured cases which is now available for general use. Integrating SAI into a CBR project eliminates the need for developing a storage scheme for graph data structures and facilitates the development of retrieval algorithms by (1) providing useful code for common retrieval tasks and (2) delineating the tasks which may require domain-specific implementations. SAI maintains flexibility through its customizability yet carries solutions for many common tasks for CBR developers, as well as illuminating some general principles for addressing common representation and retrieval needs for process-oriented CBR tasks and the relationship between CBR needs for structured retrieval and needs and methods in graph oriented database research. The paper closes with an evaluation of SAI's scalability and demonstration of the value of enabling developers to select retrieval methods suited to their tasks. It closes with a discussion of SAI's relationship to the growing body of work on graph databases. [Copyright &y& Elsevier]

Abstract: A natural metric in 2-manifold surfaces is to use geodesic distance. If a 2-manifold surface is represented by a triangle mesh , the geodesic metric on can be computed exactly using computational geometry methods. Previous work for establishing the geodesic metric on only supports using half-edge data structures; i.e., each edge in is split into two halves and each half-edge corresponds to one of two faces incident to . In this paper, we prove that the exact-geodesic structures on two half-edges of can be merged into one structure associated with . Four merits are achieved based on the properties which are studied in this paper: (1) Existing CAD systems that use edge-based data structures can directly add the geodesic distance function without changing the kernel to a half-edge data structure; (2) To find the geodesic path from inquiry points to the source, the MMP algorithm can be run in an on-the-fly fashion such that the inquiry points are covered by correct wedges; (3) The MMP algorithm is sped up by pruning unnecessary wedges during the wedge propagation process; (4) The storage of the MMP algorithm is reduced since fewer wedges need to be stored in an edge-based data structure. Experimental results show that when compared to the classic half-edge data structure, the edge-based implementation of the MMP algorithm reduces running time by 44% and storage by 29% on average. [Copyright &y& Elsevier]

Abstract: Building the behaviour for non-player characters in a game is a complex collaborative task among AI designers and programmers. In this paper we present a visual authoring tool for game designers that supports behaviour reuse. We describe a visual editor, capable of storing, indexing, retrieving and reusing behaviours previously designed by AI programmers. One of the most notable features of our editor is its capability for sketch-based retrieval: searching in a repository for behaviours that are similar to the one the user is drawing, and making suggestions about how to complete it. As this process relies on graph behaviour comparison, in this paper, we describe different algorithms for graph comparison, and demonstrate, through empirical evaluation in a particular test domain, that we can provide structure-based similarity for graphs that preserves behaviour similarity and can be computed at reasonable cost. [Copyright &y& Elsevier]

Web page ranking is one of the core components of search engines. Given a user query, ranking aims to provide a ranked list of Web pages that the user is likely to prefer the most. By and large, the ranking algorithms can be categorized into content-based approaches, link-based approaches, and hybrid approaches. Hybrid ranking algorithms, which exploit both the content and link information, are the most popular and extensively studied techniques. Among the hybrid algorithms, C-Rank combines content and link information in a very effective way using the concept of contribution. This algorithm is known to provide high performance in terms of both accurate and prompt responses to user queries. However, C-Rank suffers from very high costs to reflect the highly dynamic and extremely frequent changes in the World Wide Web, because it re-computes all of the C-Rank scores used for ranking from scratch to reflect the changes. As a result, C-Rank may be considered inappropriate to provide users with accurate and up-to-date search results. This paper aims to remedy this limitation of C-Rank. We propose incremental C-Rank , which is designed to update the C-Rank scores of only a carefully chosen portion of the Web pages rather than those of all of the Web pages without any accuracy loss. Our experimental results on a real-world dataset confirm both the effectiveness and efficiency of our proposed method. [ABSTRACT FROM AUTHOR]

Highlights • We design a semi-supervised feature representation framework for person Re-Identification which effectively utilizes both labeled and unlabeled training data to learn a discriminative representation so that person images across disjoint camera views can be reliably matched. • A multi-view clustering method is proposed to integrate features from multiple Convolutional Neural Networks for clustering, which can give more accurate label estimation for unlabeled data. • Each of our Convolutional Neural Networks utilizes a siamese network that simultaneously computes the identification loss and verification loss, which simultaneously learns a discriminative Convolutional Neural Network embedding and a similarity metric, and thus improving pedestrian retrieval accuracy. Extensive experiments on large-scale person Re-Id datasets demonstrate the effectiveness of our method. Abstract Person Re-Identification (Re-Id) is a challenging task focusing on identifying the same person among disjoint camera views. A number of deep learning algorithms have been reported for this task in fully-supervised fashion which requires a large amount of labeled training data, while obtaining high quality labels for Re-Id is extremely time consuming. To address this problem, we propose a semi-supervised Re-Id framework by using only a small portion of labeled data and some additional unlabeled samples. This paper approaches the problem by constructing a set of heterogeneous Convolutional Neural Networks (CNNs) fine-tuned using the labeled portion, and then propagating the labels to the unlabeled portion for further fine-tuning the overall system. In this work, label estimation is a key component during the propagation process. We propose a novel multi-view clustering method, which integrates features of multiple heterogeneous CNNs to cluster and generate pseudo labels for unlabeled samples. Then we fine-tune each of the multiple heterogeneous CNNs by minimizing an identification loss and a verification loss simultaneously, using training data with both true labels and pseudo labels. The procedure is iterated until the estimation of pseudo labels no longer changes. Extensive experiments on three large-scale person Re-Id datasets demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract: As the multimedia data increasing exponentially, how to get the video data we need efficiently become so important and urgent. In this paper, a novel method for shot retrieval is proposed, which is based on fuzzy evolutionary aiNet and hybrid features. To begin with, the fuzzy evolutionary aiNet algorithm proposed in this paper is utilized to extract key-frames in a video sequence. Meanwhile, to represent a key-frame, hybrid features of color feature, texture feature and spatial structure feature are extracted. Then, the features of key-frames in the same shot are taken as an ensemble and mapped to high dimension space by non-linear mapping, and the result obeys Gaussian distribution. Finally, shot similarity is measured by the probabilistic distance between distributions of the key-frame feature ensembles for two shots, and similar shots are retrieved effectively by using this method. Experimental results show the validity of this proposed method. [Copyright &y& Elsevier]

Abstract: This paper proposes a novel method for distributed data organization and parallel data retrieval from huge volume point clouds generated by airborne Light Detection and Ranging (LiDAR) technology under a cluster computing environment, in order to allow fast analysis, processing, and visualization of the point clouds within a given area. The proposed method is suitable for both grid and quadtree data structures. As for distribution strategy, cross distribution of the dataset would be more efficient than serial distribution in terms of non-redundant datasets, since a dataset is more uniformly distributed in the former arrangement. However, redundant datasets are necessary in order to meet the frequent need of input and output operations in multi-client scenarios: the first copy would be distributed by a cross distribution strategy while the second (and later) would be distributed by an iterated exchanging distribution strategy. Such a distribution strategy would distribute datasets more uniformly to each data server. In data retrieval, a greedy algorithm is used to allocate the query task to a data server, where the computing load is lightest if the data block needing to be retrieved is stored among multiple data servers. Experiments show that the method proposed in this paper can satisfy the demands of frequent and fast data query. [ABSTRACT FROM AUTHOR]

Abstract: This paper investigates feature selection based on rough sets for dimensionality reduction in Case-Based Reasoning classifiers. In order to be useful, Case-Based Reasoning systems should be able to manage imprecise, uncertain and redundant data to retrieve the most relevant information in a potentially overwhelming quantity of data. Rough Set Theory has been shown to be an effective tool for data mining and for uncertainty management. This paper has two central contributions: (1) it develops three strategies for feature selection, and (2) it proposes several measures for estimating attribute relevance based on Rough Set Theory. Although we concentrate on Case-Based Reasoning classifiers, the proposals are general enough to be applicable to a wide range of learning algorithms. We applied these proposals on twenty data sets from the UCI repository and examined the impact of feature selection over classification performance. Our evaluation shows that all three proposals benefit the basic Case-Based Reasoning system. They also present robustness in comparison to well-known feature selection strategies. [ABSTRACT FROM AUTHOR]

Abstract: Novelty detection aims at reducing redundant information from a chronologically ordered list of documents or sentences. Other studies of novelty detection have been conducted on the English language, but few papers have addressed the problem of multilingual novelty detection. Likewise, research in multilingual information retrieval have rarely been applied to novelty detection. This paper attempts to bridge the two disciplines by first describing the preprocessing steps for English, Malay and Chinese, then applying document and sentence-level novelty detection for the three languages on APWSJ and TREC 2004 Novelty Track data. Experiments on sentence-level novelty detection show similar results for all three languages, which indicates that our algorithm is suitable for multilingual novelty detection at the sentence level. However, results for document-level novelty detection show a disparity across the different languages, with English and Malay outperforming Chinese. After applying sentence-level novelty detection to detect novel documents, we observe substantial improvements on all three languages. This demonstrates that segmenting documents into sentences improves document-level novelty detection in multiple languages, and has practical benefits for a real-time multilingual novelty detection system. [Copyright &y& Elsevier]

Transforming unstructured text into a formal representation is an important goal of the Semantic Web in order to facilitate the integration and retrieval of information. The construction of Knowledge Graphs (KGs) pursues such an idea, where named entities (real world things) and their relations are extracted from text. In recent years, many approaches for the construction of KGs have been proposed by exploiting Discourse Analysis, Semantic Frames, or Machine Learning algorithms with existing Semantic Web data. Although such approaches are useful for processing taxonomies and connecting beliefs, they provide several linguistic descriptions, which lead to semantic data heterogeneity and thus, complicating data consumption. Moreover, Open Information Extraction (OpenIE) approaches have been slightly explored for the construction of KGs, which provide binary relations representing atomic units of information that could simplify the querying and representation of data. In this paper, we propose an approach to generate KGs using binary relations produced by an OpenIE approach. For such purpose, we present strategies for favoring the extraction and linking of named entities with KG individuals, and additionally, their association with grammatical units that lead to producing more coherent facts. We also provide decisions for selecting the extracted information elements for creating potentially useful RDF triples for the KG. Our results demonstrate that the integration of information extraction units with grammatical structures provides a better understanding of proposition-based representations provided by OpenIE for supporting the construction of KGs. [ABSTRACT FROM AUTHOR]

Clustering is an important technique to deal with large scale data which are explosively created in internet. Most data are high-dimensional with a lot of noise, which brings great challenges to retrieval, classification and understanding. No current existing approach is “optimal” for large scale data. For example, DBSCAN requires O ( n 2 ) time, Fast-DBSCAN only works well in 2 dimensions, and ρ -Approximate DBSCAN runs in O ( n ) expected time which needs dimension D to be a relative small constant for the linear running time to hold. However, we prove theoretically and experimentally that ρ -Approximate DBSCAN degenerates to an O ( n 2 ) algorithm in very high dimension such that 2 D  >  >  n . In this paper, we propose a novel local neighborhood searching technique, and apply it to improve DBSCAN, named as NQ-DBSCAN, such that a large number of unnecessary distance computations can be effectively reduced. Theoretical analysis and experimental results show that NQ-DBSCAN averagely runs in O ( n * log ( n )) with the help of indexing technique, and the best case is O ( n ) if proper parameters are used, which makes it suitable for many realtime data. [ABSTRACT FROM AUTHOR]

Abstract: Web Usage Mining (WUM) is the application of data mining techniques over Web server logs in order to extract navigation usage patterns. The analysis of mining patterns for assessing the knowledge they reveal is a critical phase in WUM. The main challenges are: (a) mining algorithms yield a huge number of patterns and (b) there is a significant semantic gap between URLs and events performed by users. In this paper, we describe the pattern analysis mechanisms integrated in O3R (Ontology-based Rules Retrieval and Rummaging), a human-centered environment for the analysis of navigation rules. O3R explores the synergy of mechanisms for retrieving and analyzing patterns. Filtering and clustering allow users to retrieve subsets of patterns with specific characteristics, in order to deal with the large volume of patterns. Rummaging mechanisms are targeted at assessing the meaning and relevance of pattern with regard to the domain, and it is particularly suitable for exploratory analysis. The distinctive feature of O3R is that is dynamically associates meaning to patterns using the concepts and relationships of a domain ontology, as a means of reducing the gap between syntactic URLs and semantic events performed by users. The paper describes the mechanisms in detail, and explores their synergic integration in the O3R prototype. It also reports two case studies that evaluate the use of O3R for the analysis of navigation patterns of a learning site. [Copyright &y& Elsevier]

Compressing symmetrical or quasi-symmetrical information can decrease data redundancy, save storage and network resources, and speed up the real-time transmission of symmetrical data over a network. During the past few decades, the compression and decompression of a symmetrical matrix, the index transform formulae have been used to calculate the post-compression indices from the original indices. However, the index inverse transform formulae that are used to calculate the original indices from the post-compression indices have yet to be derived. The algorithms used to decompress these matrices have never been very efficient, such as the bi-layer loops algorithm, which is proportional to the time complexity factor O(n 2). Using the index inverse transform formulae deduced in this paper, the corresponding time complexity factor can be reduced to O(1), which is a significant improvement. This paper analyses the difficulty in deriving an index inverse transform formula. The formulae for the index transform and its inverse transform of the column-compression of the symmetrical matrix are proposed and can be stated as five theorems. Based on those formulae, the relative running efficiency of two decompressions on the covariance matrix of the positions and velocities of ITRF2000—ALASKA Stations is analyzed. It is concluded that two redundant fields in the International Terrestrial Reference Frame (ITRF) covariance matrix file can be removed. Other related issues for further study in the geosciences are also raised. [Copyright &y& Elsevier]

Recently, the survival signature has been presented as a summary of the structure function which is sufficient for computation of common reliability metrics and has the crucial advantage that it can be applied to systems with components whose failure times are not exchangeable. The survival signature provides a huge reduction in required information, e.g. for its storage, compared to the full structure function, its implementation to larger systems is still difficult in a purely analytical manner and simulations may be required to derive the reliability metrics of interest. Hence, the main question addressed in this paper is whether or not the survival signature provides sufficient information for efficient simulation to derive the system’s failure time distribution. We answer this question in the affirmative by presenting two algorithms for survival signature-based simulation. In addition, we present a third simulation algorithm that can be used in case of repairable components. It turns out that these algorithms are very efficient, beyond the initial advantage of requiring only the survival signature to be available, instead of the full structure function. [ABSTRACT FROM AUTHOR]

Automatic Vehicles Counting and Recognizing (AVCR) is a very challenging topic in transport engineering having important implications for the modern transport policies. Implementing a computer-assisted AVCR in the most vital districts of a country provides a large amount of measurements which are statistically processed and analyzed, the purpose of which is to optimize the decision-making of traffic operation, pavement design, and transportation planning. Since, the advent of computer vision technology, video-based surveillance of road vehicles has become a key component in developing autonomous intelligent transportation systems. In this context, this paper proposes a Pattern Recognition system which employs an unsupervised clustering algorithm with the objective of detecting, counting and recognizing a number of dynamic objects crossing a roadway. This strategy defines a virtual sensor, whose aim is similar to that of an inductive-loop in a traditional mechanism, i.e. to extract from the traffic video streaming a number of signals containing anarchic information about the road traffic. Then, the set of signals is filtered with the aim of conserving only motion’s significant patterns. Resulted data are subsequently processed by a statistical analysis technique so as to estimate and try to recognize a number of clusters corresponding to vehicles. Finite mixture models fitted by the EM algorithm are used to assess such clusters, which provides important information about the traffic such as the instantaneous number of vehicles, their weights, velocities and intervehicular distances. [ABSTRACT FROM AUTHOR]

Topic modeling is used in information retrieval to infer the hidden themes in a collection of documents and thus provides an automatic means to organize, understand and summarize large collections of textual information. Topic models also offer an interpretable representation of documents used in several downstream Natural Language Processing (NLP) tasks. Modeling techniques vary from probabilistic graphical models to the more recent neural models. This paper surveys topic models from four aspects. The first aspect categorizes different topic modeling techniques into four categories: algebraic, fuzzy, probabilistic, and neural. We review the wide variety of available models from each category, highlight differences and similarities between models and model categories using a unified perspective, investigate these models' characteristics and limitations, and discuss their proper use cases. The second aspect illustrates six criteria for proper evaluation of topic models, from modeling quality to interpretability, stability, efficiency, and beyond. Topic modeling has found applications in various disciplines, owing to its interpretability. We examine these applications along with some popular software tools which provide an implementation of some models. The fourth aspect reviews available datasets and benchmarks. Using two benchmark datasets, we conducted experiments to compare seven topic models along the proposed metrics. The discussion highlights the differences between the models and their relative suitability for various applications. It notes the relationship between evaluation metrics and proposes four key aspects to help decide which model to use for an application. Our discussion also shows that the research trends move towards developing and tuning neural topic models and leveraging the power of pre-trained language models. Finally, it highlights research gaps in developing unified benchmarks and evaluation metrics. • Categorization of topic models into algebraic, probabilistic, fuzzy and neural. • Surveys models' evaluation criteria and different metrics used. • Surveys topic modeling software and applications. • Proposes a method to help select a topic model for an application. • Compares topic models across coherence, diversity, stability and efficiency. [ABSTRACT FROM AUTHOR]

Efficient and reliable slip sensor plays a critical role in grasp of deformable object. This paper provides a new method for slip detection based on video processing, which can offer multidimensional sensing information and it is applicable for gripping deformable objects. The macro camera mounted inside the gripper can capture the dynamic slip and distortion of the object. The features of interest points, detected by the SURF algorithm in subsequent frames, are used to identify the properties of object and calculate the object’s changes. To determine the object’s slippage, nine detecting windows in the camera view can detect nine tracks of the slippage. Besides, the object’s distortion and direction of tangential force on the surface can be determined from these tracks, which helps to the estimation of the object’s damage. Experiments on homogeneous and textured materials show that this method can detect the object’ slippage accurately, and meanwhile provides the tracks of the slippage, distortion information and the force situation. Furthermore, tests for comparison were carried out and the results show that this sensor has a high accuracy in detecting slippage and distortion and the error is acceptable, which has an significant impact in deformable object grasp. [ABSTRACT FROM AUTHOR]

Era of knowledge economy, how to effectively mining, the use of knowledge is the enterprise growing concern. CBR system from the field of artificial intelligence is a self-learning system to manage tacit knowledge (case). Case retrieval link is the core link, the advantages and disadvantages of search methods directly affect the efficiency of case retrieval and case matching accuracy. Therefore, this paper proposes a new case matching process: when the size of the case database is small, it searches based on the case similarity algorithm; when the case database is large, it searches based on the FCM secondary retrieval model. And illustrates the fastness and efficiency of FCM in matching large-scale case database. [ABSTRACT FROM AUTHOR]

The Extensible Markup Language (XML) data model has recently gained huge popularity because of its ability to represent a wide variety of structured (relational) and semi-structured (document) data. Several query languages have been proposed for the XML model, the most-widely used one is the XQuery. An important component of an XQuery is its XPath expression which retrieves a set of XML documents to be manipulated by the associated XQuery. An XPath expression can be of several types, among which are the containment queries. Traditional research of processing containment queries has concentrated on data retrieval from independent XML documents; not much research has been directed towards interlinked XML documents. This paper reviews this area of research and shows the adequacy and correctness of one of the reviewed algorithms when applied to independent XML documents. However, the direct application of this algorithm to process queries against interlinked XML documents is shown to generate incorrect results. To remedy such a situation, two new algorithms and the associated indexing structures are developed and shown to perform correctly in processing both independent and/or inter-linked XML documents. In addition, one of the new algorithms is shown to minimize the storage requirement of the intermediate lists generated throughout its execution and therefore improving further the algorithm's space and time performance. [ABSTRACT FROM AUTHOR]

Currently, the lifestyle of elderly people is regularly monitored in order to establish guidelines for rehabilitation processes or ensure the welfare of this segment of the population. In this sense, activity recognition is essential to detect an objective set of behaviors throughout the day. This paper describes an accurate, comfortable and efficient system, which monitors the physical activity carried out by the user. An extension to an awarded activity recognition system that participated in the EvAAL 2012 and EvAAL 2013 competitions is presented. This approach uses data retrieved from accelerometer sensors to generate discrete variables and it is tested in a non-controlled environment. In order to achieve the goal, the core of the algorithm Ameva is used to develop an innovative selection, discretization and classification technique for activity recognition. Moreover, with the purpose of reducing the cost and increasing user acceptance and usability, the entire system uses only a smartphone to recover all the information required. [ABSTRACT FROM AUTHOR]

Increased adoption and deployment of phasor measurement units (PMU) has provided valuable fine-grained data over the grid. Analysis over these data can provide insight into the health of the grid, thereby improving control over operations. Realizing this data-driven control, however, requires validating, processing and storing massive amounts of PMU data. This paper describes a PMU data management system that supports input from multiple PMU data streams, features an event-detection algorithm, and provides an efficient method for retrieving archival data. The event-detection algorithm rapidly correlates multiple PMU data streams, providing details on events occurring within the power system. The event-detection algorithm feeds into a visualization component, allowing operators to recognize events as they occur. The indexing and data retrieval mechanism facilitates fast access to archived PMU data. Using this method, we achieved over 30× speedup for queries with high selectivity. With the development of these two components, we have developed a system that allows efficient analysis of multiple time-aligned PMU data streams. [ABSTRACT FROM AUTHOR]

One of the important tasks in relational data analysis is link prediction which has been successfully applied on many applications such as bioinformatics, information retrieval, etc. The link prediction is defined as predicting the existence or absence of edges between nodes of a network. In this paper, we propose a novel method for link prediction based on Distance Dependent Chinese Restaurant Process (DDCRP) model which enables us to utilize the information of the topological structure of the network such as shortest path and connectivity of the nodes. We also propose a new Gibbs sampling algorithm for computing the posterior distribution of the hidden variables based on the training data. Experimental results on three real-world datasets show the superiority of the proposed method over other probabilistic models for link prediction problem. [ABSTRACT FROM AUTHOR]

Rank aggregation mechanisms have been used in solving problems from various domains such as bioinformatics, natural language processing, information retrieval, etc. Metasearch is one such application where a user gives a query to the metasearch engine, and the metasearch engine forwards the query to multiple individual search engines. Results or rankings returned by these individual search engines are combined using rank aggregation algorithms to produce the final result to be displayed to the user. We identify few aspects that should be kept in mind for designing any rank aggregation algorithm for metasearch. For example, generally equal importance is given to the input rankings while performing the aggregation. However, depending on the indexed set of web pages, features considered for ranking, ranking functions used etc. by the individual search engines, the individual rankings may be of different qualities. So, the aggregation algorithm should give more weight to the better rankings while giving less weight to others. Also, since the aggregation is performed when the user is waiting for response, the operations performed in the algorithm need to be light weight. Moreover, getting supervised data for rank aggregation problem is often difficult. In this paper, we present an unsupervised rank aggregation algorithm that is suitable for metasearch and addresses the aspects mentioned above. We also perform detailed experimental evaluation of the proposed algorithm on four different benchmark datasets having ground truth information. Apart from the unsupervised Kendall-Tau distance measure, several supervised evaluation measures are used for performance comparison. Experimental results demonstrate the efficacy of the proposed algorithm over baseline methods in terms of supervised evaluation metrics. Through these experiments we also show that Kendall-Tau distance metric may not be suitable for evaluating rank aggregation algorithms for metasearch. [ABSTRACT FROM AUTHOR]

Advances in computational power have enabled the manipulation of music audio files and the emergence of the Music Information Retrieval field. One of the main research lines in this area is that of Music Transcription , which aims at transforming an audio file into musical notation. So far, most efforts have focused on accurately transcribing the pitch and durations of the notes, and thus neglecting other aspects in the music score. The present work explores a novel line of action in the context of automatic music transcription, focusing on the dynamics, and by means of Linguistic Description . The process described in this paper (called MUDELD : MU sic D ynamics E xtraction through L inguistic D escription) departs from the data series representing the audio file, and requires the segmentation of the piece in phrases, which is currently done by hand. Initial experiments have been performed on eight recordings of Debussy's Syrinx with promising results. [ABSTRACT FROM AUTHOR]

This paper proposes a distinct multiscale shape context (MSC) method for isometric 3D shape description and retrieval. For each feature point, a MSC descriptor is devised to capture the multiple spatial information on the basis of the intrinsic shape context, which is advantageous in solving the domain offset deficiency for intra-class shapes. Different from the traditional shape context method, the MSC descriptor is built based on the charts without angular bins and the shape distributions in local domains, which makes it not only simple but also efficient. To reduce the cost of shape representation, we detect a sparse set of feature points and design an improved bags-of-word model to encode the MSC descriptors. For retrieval improvement, an efficient while robust re-ranking algorithm by metric mapping is designed to alleviate the errors of the feature space. Finally, the experimental results have demonstrated significant performance gains on two public benchmarks. [ABSTRACT FROM AUTHOR]

In dynamic information environments such as the web, the amount of information is rapidly increasing. Thus, the need to organize such information in an efficient manner is more important than ever. With such dynamic nature, incremental clustering algorithms are always preferred compared to traditional static algorithms. In this paper, an enhanced version of the incremental DBSCAN algorithm is introduced for incrementally building and updating arbitrary shaped clusters in large datasets. The proposed algorithm enhances the incremental clustering process by limiting the search space to partitions rather than the whole dataset which results in significant improvements in the performance compared to relevant incremental clustering algorithms. Experimental results with datasets of different sizes and dimensions show that the proposed algorithm speeds up the incremental clustering process by factor up to 3.2 compared to existing incremental algorithms. [ABSTRACT FROM AUTHOR]

Instance selection aims at filtering out noisy data (or outliers) from a given training set, which not only reduces the need for storage space, but can also ensure that the classifier trained by the reduced set provides similar or better performance than the baseline classifier trained by the original set. However, since there are numerous instance selection algorithms, there is no concrete winner that is the best for various problem domain datasets. In other words, the instance selection performance is algorithm and dataset dependent. One main reason for this is because it is very hard to define what the outliers are over different datasets. It should be noted that, using a specific instance selection algorithm, over-selection may occur by filtering out too many ‘good’ data samples, which leads to the classifier providing worse performance than the baseline. In this paper, we introduce a dual classification (DuC) approach, which aims to deal with the potential drawback of over-selection. Specifically, performing instance selection over a given training set, two classifiers are trained using both a ‘good’ and ‘noisy’ sets respectively identified by the instance selection algorithm. Then, a test sample is used to compare the similarities between the data in the good and noisy sets. This comparison guides the input of the test sample to one of the two classifiers. The experiments are conducted using 50 small scale and 4 large scale datasets and the results demonstrate the superior performance of the proposed DuC approach over the baseline instance selection approach. [ABSTRACT FROM AUTHOR]

Objective: Case-based reasoning (CBR) is a problem-solving paradigm that uses past knowledge to interpret or solve new problems. It is suitable for experience-based and theory-less problems. Building a semantically intelligent CBR that mimic the expert thinking can solve many problems especially medical ones.Methods: Knowledge-intensive CBR using formal ontologies is an evolvement of this paradigm. Ontologies can be used for case representation and storage, and it can be used as a background knowledge. Using standard medical ontologies, such as SNOMED CT, enhances the interoperability and integration with the health care systems. Moreover, utilizing vague or imprecise knowledge further improves the CBR semantic effectiveness. This paper proposes a fuzzy ontology-based CBR framework. It proposes a fuzzy case-base OWL2 ontology, and a fuzzy semantic retrieval algorithm that handles many feature types.Material: This framework is implemented and tested on the diabetes diagnosis problem. The fuzzy ontology is populated with 60 real diabetic cases. The effectiveness of the proposed approach is illustrated with a set of experiments and case studies.Results: The resulting system can answer complex medical queries related to semantic understanding of medical concepts and handling of vague terms. The resulting fuzzy case-base ontology has 63 concepts, 54 (fuzzy) object properties, 138 (fuzzy) datatype properties, 105 fuzzy datatypes, and 2640 instances. The system achieves an accuracy of 97.67%. We compare our framework with existing CBR systems and a set of five machine-learning classifiers; our system outperforms all of these systems.Conclusion: Building an integrated CBR system can improve its performance. Representing CBR knowledge using the fuzzy ontology and building a case retrieval algorithm that treats different features differently improves the accuracy of the resulting systems. [ABSTRACT FROM AUTHOR]

In this research paper, we propose novel features based on information theory for image retrieval. We propose the novel concept of “probabilistic filtering”. We propose a hybrid approach for image retrieval that combines annotation approach with content based image retrieval approach. Also rough set theory is proposed as a tool for audio/video object retrieval from multi-media databases. [ABSTRACT FROM AUTHOR]

Arabic dialects arewidely used from many years ago instead of Modern Standard Arabic language in many fields. The presence of dialects in any language is a big challenge. Dialects add a new set of variational dimensions in some fields like natural language processing, information retrieval and even in Arabic chatting between different Arab nationals. Spoken dialects have no standard morphological, phonological and lexical like Modern Standard Arabic. Hence, the objective of this paper is to describe a procedure or algorithm by which a stem for the Arabian Gulf dialect can be defined. The algorithm is rule based. Special rules are created to remove the suffixes and prefixes of the dialect words. Also, the algorithm applies rules related to the word size and the relation between adjacent letters. The algorithm was tested for a number of words and given a good correct stem ratio. The algorithm is also compared with two Modern Standard Arabic algorithms. The results showed that Modern Standard Arabic stemmers performed poorly with Arabic Gulf dialect and our algorithm performed poorly when applied for Modern Standard Arabic words. [ABSTRACT FROM AUTHOR]

Stemming algorithms are used in information retrieval systems, indexers, text mining, text classifiers etc., to extract stems or roots of different words, so that words derived from the same stem or root are grouped together. Many stemming algorithms were built in different natural languages. Khoja stemmer is one of the known and widely used Arabic stemmers. In this paper, we introduced a new light and heavy Arabic stemmer. This new stemmer is presented in this study and compared with two well-known Arabic stemmers. Results showed that accuracy of our stemmer is slightly better than the accuracy yielded by each one of those two well-known Arabic stemmers used for evaluation and comparison. Evaluation tests on our novel stemmer yield 75.03% accuracy, while the other two Arabic stemmers yield slightly lower accuracy. [ABSTRACT FROM AUTHOR]

Finding the most influential people is an NP-hard problem that has attracted many researchers in the field of social networks. The problem is also known as influence maximization and aims to find a number of people that are able to maximize the spread of influence through a target social network. In this paper, a new algorithm based on the linear threshold model of influence maximization is proposed. The main benefit of the algorithm is that it reduces the number of investigated nodes without loss of quality to decrease its execution time. Our experimental results based on two well-known datasets show that the proposed algorithm is much faster and at the same time more efficient than the state of the art algorithms. [ABSTRACT FROM AUTHOR]

With the popularization of media-capture devices and the development of the Internet's basic facilities, video has become the most popular media information in recent years. The massive capacity of video imposes the demand of automatic video identification techniques which are very important to various applications such as content based video retrieval and copy detection. Therefore, as a challenging problem, video identification has drawn more and more attention in the past decade. The problem addressed here is to identify a given video clip in a given set of video sequences. In this paper, a robust video identification algorithm based on local non-negative matrix factorization (LNMF) is presented. First, some concepts about LNMF are described and the way of finding the factorized matrix is given. Then, its convergence is proven. In addition, a LNMF based shot detection method is proposed for constructing a video identification framework completely based on LNMF. Finally, a LNMF based identification approach using Hausdorff distance is introduced and a two-stage search process is proposed. Experimental results show the robustness of the proposed approach to many kinds of content-preserved distortions and its superiority to other algorithms. [ABSTRACT FROM AUTHOR]

Similarity learning plays a fundamental role in the problem of database retrieval and nearest classification problem. Traditional pairwise similarity measure ignores the contextual information, and the Graph Transduction (GT) has been proposed as a contextual similarity learning algorithm to utilize the contextual information, which is embedded in a nearest neighbor graph. On main shortage of this method is that it is difficult to choose the optimal graph since different graphs may focus on different aspects of the objects. Co-Transduction (CT) is lately proposed by fusing two different graphs. In this paper, we generalize this problem by using the ensemble of many candidate graphs with different models and parameters for transduction, by assuming that the optimal graph could be obtained by the weighted linear ensemble of these candidate graphs. The similarities and graph weights are modeled within one unified objective function, and optimized alternately in an iterative algorithm. The new proposed algorithm, named as Ensemble Transduction (ET), is tested on two challenging tasks and the experimental results show that it can outperform both the GT and CT. [ABSTRACT FROM AUTHOR]

Abstract: Estimation of additive driving-forces (e.g., hidden inputs) in nonlinear dynamic systems is addressed with varying amounts of a priori knowledge on system models exemplified by three typical scenarios: (1) there is no sufficient prior knowledge to build a mathematical model of the underlying system; (2) the system is partially described by an analytic model; (3) a complete and accurate model of the underlying system is available. Three algorithms are proposed for each scenario and analyzed comprehensively. The adaptive driving-force estimator (ADFE) [1,2] is used for the retrieval of driving-forces using only the system outputs for the first scenario. A variational Bayesian and a Bayesian algorithm are established for the second and the third scenarios, respectively. All three algorithms are studied in depth on a nonlinear dynamic system with equivalent computational resources, and the Posterior Cramer–Rao Lower Bounds (PCRLB) are specified as performance metrics for each case. The results lead to a thorough understanding of the capabilities and limitations of the ADFE, which manifests itself as an effective technique for the estimation of rapidly varying hidden inputs unless a complete and accurate model is available. Moreover, the methods developed in this paper facilitate a suitable framework for the construction of new and efficient tools for various input estimation problems. In particular, the proposed algorithms constitute a readily available basis for the design of novel input residual estimators to approach the Fault Diagnosis and Isolation (FDI) problem from a new and different perspective. [Copyright &y& Elsevier]

Abstract: Semantic gap has become a bottleneck of content-based image retrieval in recent years. In order to bridge the gap and improve the retrieval performance, automatic image annotation has emerged as a crucial problem. In this paper, a hybrid approach is proposed to learn the semantic concepts of images automatically. Firstly, we present continuous probabilistic latent semantic analysis (PLSA) and derive its corresponding Expectation–Maximization (EM) algorithm. Continuous PLSA assumes that elements are sampled from a multivariate Gaussian distribution given a latent aspect, instead of a multinomial one in traditional PLSA. Furthermore, we propose a hybrid framework which employs continuous PLSA to model visual features of images in generative learning stage and uses ensembles of classifier chains to classify the multi-label data in discriminative learning stage. Therefore, the framework can learn the correlations between features as well as the correlations between words. Since the hybrid approach combines the advantages of generative and discriminative learning, it can predict semantic annotation precisely for unseen images. Finally, we conduct the experiments on three baseline datasets and the results show that our approach outperforms many state-of-the-art approaches. [Copyright &y& Elsevier]