Similarity Geometry

I would like to acknowledge John J. Borg for the guidance throughout the project and the permission to use the birds from the Natural History Museum. I would also like to thank my advisor Dr. Louis Cassar for supporting my project. I would like to specially thank Fran Pisani for all the help in the lab and Dr. Belinda Gambin for allowing me use of the lab and its instruments. I would like to thank my program advisors, Dr. Maria Papadakis and Dr. Liz Conrad for giving me the opportunity to write a thesis and for all of their support throughout the entire program. I would also like to thank Dr. Amy Goodall for serving on my committee and for all the revision help. Finally, I would like to thank all the people who supported me throughout the entire process and helped me even with the dirty work.

Through this article, we aim to introduce Holos-a new collaborative environment that allows researchers to carry out experiments based on similarity assessments between stimuli, such as in projective-mapping and sorting tasks. An important feature of Holos is its capacity to assess real-time individual processes during the task. Within the Holos environment, researchers can design experiments on its platform, which can handle four kinds of stimuli: concepts, images, sounds, and videos. In addition, researchers can share their study resources within the scientific community, including stimuli, experimental protocols, and/or the data collected. With a dedicated Android application combined with a tactile human-machine interface, subjects can perform experiments using a tablet to obtain similarity measures between stimuli. On the tablet, the stimuli are displayed as icons that can be dragged with one finger to position them, depending on the ways they are perceived. By recording the x,y coordinates of the stimuli while subjects move the icons, the obtained data can reveal the cognitive processes of the subjects during the experiment. Such data, named digit-tracking data, can be analyzed with the SensoMineR package. In this article, we describe how researchers can design an experiment, how subjects can perform the experiment, and how digit-tracking data can be statistically analyzed within the Holos environment. At the end of the article, a short exemplary experiment is presented.

Abstract. In this paper, we propose a new approach to generate analogy questions of the form” A is to B as... is to?” from ontologies. Analogy questions are widely used in multiple-choice tests such as SATs and GREs and are used to assess student's higher cognitive abilities. The design, implementation and evaluation of the new approach are presented in this paper. The results show that mining ontologies for such questions is fruitful.

We address the problem of analysing the joint coherence of a number of concepts with respect to a background ontology. To address this problem, we explore the applicability of Paul Thagard’s computational theory of coherence, in combination with semantic similarity between concepts based on a generalisation operator. In particular, given the input concepts, our approach computes maximally coherent subsets of these concepts following Thagard’s partitioning approach, whilst returning a number of possible generalisations of these concepts as justification of why these concepts cohere.

This paper describes the specifications and results of UMCC_DLSI system, which participated in the first Semantic Textual Similarity task (STS) of SemEval-2012. Our supervised system uses different kinds of semantic and lexical features to train classifiers and it uses a voting process to select the correct option. Related to the different features we can highlight the resource ISR-WN 1 used to extract semantic relations among words and the use of different algorithms to establish semantic and lexical similarities. In order to establish which features are the most appropriate to improve STS results we participated with three runs using different set of features. Our best approach reached the position 18 of 89 runs, obtaining a general correlation coefficient up to 0.72.

The use of semantics in tasks related to information retrieval has become, in recent years, a vast field of research. Considering supervised text classification, which is the main interest of this work, semantics can be involved at different steps of text processing: during indexing step, during training step and during class prediction step. As for class prediction step, semantics can be involved through text-to-text semantic similarity measures. These measures can replace classical similarity measures that are traditionally used by some classification methods for decision-making. In this paper we propose a new measure for assessing semantic similarity between texts based on TF/IDF with a new function that aggregates semantic similarities between concepts representing the compared text documents pair-to-pair using a semantic similarity matrix. Considering five different ontology-based semantic similarity measures, we experimentally evaluate this new text-to-text semantic similarity measure in the context of supervised medical text classification using Ohsumed corpus. Experimental results demonstrate that our measure outperforms other semantic and classical measures with significant improvements in the concept space.

Superspirals include a very broad family of monotonic curvature curves, whose radius of curvature is defined by a completely monotonic Gauss hypergeometric function. They are generalizations of log-aesthetic curves, and other curves whose radius of curvature is a particular case of a completely monotonic Gauss hypergeometric function. In this work, we study superspirals of confluent type via similarity geometry. Through a detailed investigation of the similarity curvatures of superspirals of confluent type, we find a new class of planar curves with monotone curvature in terms of Tricomi confluent hypergeometric function. Moreover, the proposed ideas will be our guide to expanding superspirals.

This paper proposes an interactive approach for region-based image clustering and retrieval. By performing clustering before image retrieval, the search space can be reduced to those clusters that are close to the query target. First, the image is segmented to regions by using an unsupervised segmentation method. This is an area where a vast number of regions are involved. To reduce search space for region-based image retrieval, we use clustering based on genetic algorithm. Fuzzy similarity is used in order to compute the similarity of two images. Moreover, a two-class SVM is trained based on user interests to improve image retrieval. Experiments were performed on COREL image database and show the effectiveness of the proposed approach.

This paper presents a method to discover initial global similarity weights while developing a case-based reasoning (CBR) system. The approach is based on multiple feature relevance scoring methods and the relevance of features within each scoring method. The objective of this work is to utilize the characteristics of a dataset when creating similarity measures. The primary advantage of this method lies in its data-driven approach in the absence of domain knowledge in the early phase of a CBR system development. The results obtained based on the experiments on multiple public datasets show that the method improves the performance of similarity measures for a CBR system in discriminating relevant similar cases. Evaluation of the results is based on the method suitable for unbalanced datasets.

We present the Penn system for SemEval-2012 Task 6, computing the degree of semantic equivalence between two sentences. We explore the contributions of different vector models for computing sentence and word similarity: Collobert and Weston embeddings as well as two novel approaches, namely eigenwords and selectors. These embeddings provide different measures of distributional similarity between words, and their contexts. We used regression to combine the different similarity measures, and found that ...

A paint roller assembly includes three paint rollers mounted in a mutually parallel relation on a support frame. Two of the rollers are spaced from each other, and are positioned on the frame for rolling engagement with a surface to be coated with paint. A third roller is carried in rolling engagement with the other two rollers for preventing splatter of paint from the rollers, and for distributing paint between the rollers. A handle member is pivotally joined to the support frame by a friction clutch assembly for applying an adjustable frictional force to resist pivoting motion of the handle member.

Nowadays, there is an increasing development of intelligent systems like online social networks, personalized recommendation systems and knowledge-based systems which are especially based on ontologies. Personalized recommendation systems applied with online social networking assist delivering a personalized content to Web-based application users. Indeed, these systems offer services that can greatly improve the response to users' needs in their search for persons or for some products. In order to model these users, semantic web technologies such as ontologies are used to explicit the hidden knowledge through using rules. In this paper, we propose to measure the similarity between the user context and other users' contexts in our ontology. Then, we integrate this measure in recommendation model to infer recommendation items (raw material, production tool, supplier name, etc.) based on SWRL rules. The experiments and evaluations show the applicability of our approach.

Many developments have taken place in the field of face-recognition and liveness analysis to improvise various device securities and attendance verification systems. Many approaches have incorporated 3D analysis of the face to predict the liveness of the person in front of it. Our method tries to account for this problem without using advanced 3D imaging techniques or hardware. This results in a solution that is both, more economical and also much easier to deploy. It consists of two parts; the former helps in face verification and the latter to check the liveness of the face. In the first part, we have used a model based on Google’s FaceNet Model which learns a mapping from face images to compact Euclidean space distances, which directly correspond to the measure of similarity of the images. Once the space has been produced, face verification can be easily implemented using standard techniques with embeddings as feature vectors. For the second part, we have employed a cascaded multi-task framework that extracts certain features from the facial image which are then used to check for liveness by tracking their relative displacements. These extracted features were used to check the liveness of the person’s face by asking them to perform some tasks in a random order like head and facial movements etc.

Spreadsheets are by far the most prominent example of end-user programs of ample size and substantial structural complexity. They are usually not thoroughly tested so they often contain faults. Debugging spreadsheets is a hard task due to the size and structure, which is usually not directly visible to the user, i.e., the functions are hidden and only the computed values are presented. A way to locate faulty cells in spreadsheets is by adapting software debugging approaches for traditional procedural or object-oriented programming languages. One of such approaches is spectrum-based fault localization (Sfl). In this paper, we study the impact of different similarity coefficients on the accuracy of Sfl applied to the spreadsheet domain. Our empirical evaluation shows that three of the 42 studied coefficients (Ochiai, Jaccard and Sorensen-Dice) require less effort by the user while inspecting the diagnostic report, and can also be used interchangeably without a loss of accuracy. In addition, we illustrate the influence of the number of correct and incorrect output cells on the diagnostic report.

The COVID-19 pandemic has led to an unprecedented challenge to public health. It resulted in global efforts to understand, record, and alleviate the disease. This research serves the purpose of generating a relevant summary related to Coronavirus. The research uses the COVID-19 Open Research Dataset (CORD-19) provided by Allen Institute for AI. The dataset contains 236,336 academic full-text articles as of July 19, 2021. This paper introduces a web-based system to handle user questions over the Coronavirus full-text scholarly articles. The system periodically runs backend services to process such large amount article with basic Natural Language Processing (NLP) techniques that include tokenization, N-Grams extraction, and part-of-speech (PoS) tagging. It automatically identifies the keywords from the question and uses cosine similarity to summarize the associated content and present to the user. This research will possibly benefit researchers, health workers as well as other individuals. Moreover, the same service can be used to train with the datasets of different domains (e.g., education) to generate a relevant summary for other user groups (e.g., students). © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Over the last two decades, radiologists have been using multi-view images to detect tumors. Computer Tomography (CT) imaging is considered as one of the reliable imaging techniques. Many medical-image-processing techniques have been developed to diagnoses lung cancer at early or later stages through CT images; however, it is still a big challenge to improve the accuracy and sensitivity of the algorithms. In this paper, we propose an algorithm based on image fusion for lung segmentation to optimize lung cancer diagnosis. The image fusion technique was developed through Laplacian Pyramid (LP) decomposition along with Adaptive Sparse Representation (ASR). The suggested fusion technique fragments medical images into different sizes using the LP. After that, the LP is used to fuse the four decomposed layers. For the evaluation purposes of the proposed technique, the Lungs Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) was used. The results showed that the Di...

In this paper, we propose two exponential similarity measures for collaborative filtering in recommender systems. The proposed similarity measures are used to estimate the distance between two users or items. Furthermore, an algorithm is proposed to use the distance obtained from similarity measures introduced in this paper to evaluate empty rating elements in the rating matrix and compare the results of these similarity measures with our exponential similarity measures. Therefore, we deduce that the algorithm using the exponential similarity measures works better than the algorithm using Manhattan, Euclidean, Jaccard, Cosine or Pearson similarity measures when taking K-Most-Similar-Users to evaluate the similarity in order to find ratings for the user that we want to predict ratings for.

Drawing upon similarity-attraction theory (Byrne, 1971), we explore whether deep-level perceived similarity or surface-level actual similarity is more predictive of agreement between supervisors and subordinates about the quality of their relationship (LMX). To examine this question we analyze supervisor-subordinate agreement on four antecedents of LMX: one surface-level similarity variable (relational demography), and three deep-level similarity variables (common perspective, receptiveness, and liking). Responses from a matched set of 203 supervisor-subordinate dyads indicate that deeplevel similarity matters more than surface-level similarity since all three deep-level perceptual variables explain variance in LMX agreement above and beyond the variance explained by surface-level actual similarity.

Development of technology in educational field brings the easier ways through the variety of facilitation for learning process, sharing files, giving assignment and assessment. Automated Essay Scoring (AES) is one of the development systems for determining a score automatically from text document source to facilitate the correction and scoring by utilizing applications that run on the computer. AES process is used to help the lecturers to score efficiently and effectively. Besides it can reduce the subjectivity scoring problem. However, implementation of AES depends on many factors and cases, such as language and mechanism of scoring process especially for essay scoring. A number of methods implemented for weighting the terms from document and reaching the solutions for handling comparative level between documents answer and expert's document still defined. In this research, we implemented the weighting of Term Frequency - Inverse Document Frequency (TF-IDF) method and Cosine Similarity with the measuring degree concept of similarity terms in a document. Tests carried out on a number of Indonesian text-based documents that have gone through the stage of pre-processing for data extraction purposes. This process results is in a ranking of the document weight that have closesness match level with expert's document.

Different binary similarity measures have been explored with different agglomerative hierarchical clustering approaches for software clustering, to make the software systems understandable and manageable. Similarity measures have strengths and weakness that results in improving and deteriorating clustering quality. Determine whether strengths of the similarity measures can be used to avoid their weaknesses for software clustering. This paper presents the strengths of some of the well known existing binary similarity measures. Using these strengths, this paper introduces an improved new binary similarity measure. A series of experiments, on five different test software systems, is presented to evaluate the effectiveness of our new binary similarity measure. The results indicate that our new measure show the combined strengths of the existing similarity measures by reducing the arbitrary decisions, increasing the number of clusters and thus improve the authoritativeness of the clustering.

Similarity and unitary equivalence can be shown to be of equivalence relations. We discuss a result showing that two similar operators have equal spectra (i.e. point and approximate point spectrum). More so, unitary equivalence results for invariant subspaces and normal operators are proved. For similar normal operators, we state the Fuglede – Putnam –Rosenblum theorem that makes proofs for similar normal operators more simplified. It is also noted that direct sums and summands are preserved under unitary equivalence. Furthermore, we show that the natural concept of equivalence between Hilbert Space operators is unitary equivalence which is stronger than similarity. By introducing the notion of quasisimilarity of operators which is the same as similarity in finite dimensional spaces, but in infinite dimensional spaces, it is a much weaker relation, we further show that quasisimilarity is an equivalence relation. We also link invariant subspaces and hyperinvariant subspaces with quasisimilarity where it is seen that similarity preserves nontrivial invariant subspaces while quasisimilarity preserves nontrivial hyperinvariant subspaces. 152 Mathematics Subject Classification: 47A10; 47A15; 47B15

Superspirals include a very broad family of monotonic curvature curves, whose radius of curvature is defined by a completely monotonic Gauss hypergeometric function. They are generalizations of log-aesthetic curves, and other curves whose radius of curvature is a particular case of a completely monotonic Gauss hypergeometric function. In this work, we study superspirals of confluent type via similarity geometry. Through a detailed investigation of the similarity curvatures of superspirals of confluent type, we find a new class of planar curves with monotone curvature in terms of Tricomi confluent hypergeometric function. Moreover, the proposed ideas will be our guide to expanding superspirals.