Search results for: similarity ranking

Authors: Guang-Ho Cha

Abstract:

This paper presents a context-sensitive media similarity search algorithm. One of the central problems regarding media search is the semantic gap between the low-level features computed automatically from media data and the human interpretation of them. This is because the notion of similarity is usually based on high-level abstraction but the low-level features do not sometimes reflect the human perception. Many media search algorithms have used the Minkowski metric to measure similarity between image pairs. However those functions cannot adequately capture the aspects of the characteristics of the human visual system as well as the nonlinear relationships in contextual information given by images in a collection. Our search algorithm tackles this problem by employing a similarity measure and a ranking strategy that reflect the nonlinearity of human perception and contextual information in a dataset. Similarity search in an image database based on this contextual information shows encouraging experimental results.

Keywords: context-sensitive search, image search, similarity ranking, similarity search

Procedia PDF Downloads 339

Authors: Subodh Chandra Shakya, Rajendra Sapkota, Aakash Tamang, Shushant Pudasaini, Sujan Adhikari, Sajjan Adhikari

Abstract:

Lots of efforts have been made in order to measure the semantic similarity between the text corpora in the documents. Techniques have been evolved to measure the similarity of two documents. One such state-of-art technique in the field of Natural Language Processing (NLP) is word to vector models, which converts the words into their word-embedding and measures the similarity between the vectors. We found this to be quite useful for the task of resume ranking. So, this research paper is the implementation of the word2vec model along with other Natural Language Processing techniques in order to rank the resumes for the particular job description so as to automate the process of hiring. The research paper proposes the system and the findings that were made during the process of building the system.

Keywords: chunking, document similarity, information extraction, natural language processing, word2vec, word embedding

Procedia PDF Downloads 131

Authors: Yogendra Sisodia

Abstract:

Researchers are becoming more interested in extracting useful information from legal documents thanks to the development of large-scale language models in natural language processing (NLP), and deep learning has accelerated the creation of powerful text mining models. Legal fields like contracts benefit greatly from semantic text search since it makes it quick and easy to find related clauses. After collecting sentence embeddings, it is relatively simple to locate sentences with a comparable meaning throughout the entire legal corpus. The author of this research investigated two pre-trained language models for this task: MiniLM and Roberta, and further fine-tuned them on Legal Contracts. The author used Multiple Negative Ranking Loss for the creation of sentence transformers. The fine-tuned language models and sentence transformers showed promising results.

Keywords: legal contracts, multiple negative ranking loss, natural language inference, sentence transformers, semantic textual similarity

Procedia PDF Downloads 72

Authors: Lidija R. Jevrić, Sanja O. Podunavac-Kuzmanović, Strahinja Z. Kovačević

Abstract:

The present study deals with the selection of the most suitable quantitative structure-retention relationship (QSRR) models which should be used in prediction of the retention behavior of basic, neutral, acidic and phenolic pesticides which belong to different classes: fungicides, herbicides, metabolites, insecticides and plant growth regulators. Sum of ranking differences (SRD) approach can give a different point of view on selection of the most consistent QSRR model. SRD approach can be applied not only for ranking of the QSRR models, but also for detection of similarity or dissimilarity among them. Applying the SRD analysis, the most similar models can be found easily. In this study, selection of the best model was carried out on the basis of the reference ranking (“golden standard”) which was defined as the row average values of logarithm of retention time (logtr) defined by high performance liquid chromatography (HPLC). Also, SRD analysis based on experimental logtr values as reference ranking revealed similar grouping of the established QSRR models already obtained by hierarchical cluster analysis (HCA).

Keywords: chemometrics, chromatography, pesticides, sum of ranking differences

Procedia PDF Downloads 350

Authors: Ahmad Shahin, Fadi Chakik, Walid Moudani

Abstract:

Our study is about elaborating a potential solution for a search engine that involves semantic technology to retrieve information and display it significantly. Semantic search engines are not used widely over the web as the majorities are still in Beta stage or under construction. Many problems face the current applications in semantic search, the major problem is to analyze and calculate the meaning of query in order to retrieve relevant information. Another problem is the ontology based index and its updates. Ranking results according to concept meaning and its relation with query is another challenge. In this paper, we are offering a light meta-engine (QESM) which uses Google search, and therefore Google’s index, with some adaptations to its returned results by adding multi-query expansion. The mission was to find a reliable ranking algorithm that involves semantics and uses concepts and meanings to rank results. At the beginning, the engine finds synonyms of each query term entered by the user based on a lexical database. Then, query expansion is applied to generate different semantically analogous sentences. These are generated randomly by combining the found synonyms and the original query terms. Our model suggests the use of semantic similarity measures between two sentences. Practically, we used this method to calculate semantic similarity between each query and the description of each page’s content generated by Google. The generated sentences are sent to Google engine one by one, and ranked again all together with the adapted ranking method (QESM). Finally, our system will place Google pages with higher similarities on the top of the results. We have conducted experimentations with 6 different queries. We have observed that most ranked results with QESM were altered with Google’s original generated pages. With our experimented queries, QESM generates frequently better accuracy than Google. In some worst cases, it behaves like Google.

Keywords: semantic search engine, Google indexing, query expansion, similarity measures

Procedia PDF Downloads 402

Authors: Elahe Sarfi, Esmat Noroozi, Farhad Hosseinzadeh Lotfi

Abstract:

One of the important issues in Data Envelopment Analysis is the ranking of Decision Making Units. In this paper, a method for ranking DMUs is presented through which the weights related to efficient units should be chosen in a way that the other units preserve a certain percentage of their efficiency with the mentioned weights. To this end, a model is presented for ranking DMUs on the base of their superefficiency by considering the mentioned restrictions related to weights. This percentage can be determined by decision Maker. If the specific percentage is unsuitable, we can find a suitable and feasible one for ranking DMUs accordingly. Furthermore, the presented model is capable of ranking all of the efficient units including nonextreme efficient ones. Finally, the presented models are utilized for two sets of data and related results are reported.

Keywords: data envelopment analysis, efficiency, ranking, weight

Procedia PDF Downloads 431

Authors: Z. Abdullah, A. R. Hamdan

Abstract:

This paper is meant to analyze the ranking of University of Malaysia Terengganu, UMT’s website in the World Wide Web. There are only few researches have been done on comparing the ranking of universities’ websites so this research will be able to determine whether the existing UMT’s website is serving its purpose which is to introduce UMT to the world. The ranking is based on hub and authority values which are accordance to the structure of the website. These values are computed using two web-searching algorithms, HITS and SALSA. Three other universities’ websites are used as the benchmarks which are UM, Harvard and Stanford. The result is clearly showing that more work has to be done on the existing UMT’s website where important pages according to the benchmarks, do not exist in UMT’s pages. The ranking of UMT’s website will act as a guideline for the web-developer to develop a more efficient website.

Keywords: algorithm, ranking, website, web structure mining

Procedia PDF Downloads 485

Authors: M. S. Annie Christi

Abstract:

Transportation Problem (TP) is based on supply and demand of commodities transported from one source to the different destinations. Usual methods for finding solution of TPs are North-West Corner Rule, Least Cost Method Vogel’s Approximation Method etc. The transportation costs tend to vary at each time. We can use fuzzy numbers which would give solution according to this situation. In this study the Best Candidate Method (BCM) is applied. For ranking Centroid Ranking Technique (CRT) and Robust Ranking Technique have been adopted to transform the fuzzy TP and the above methods are applied to EDWARDS Vacuum Company, Crawley, in West Sussex in the United Kingdom. A Comparative study is also given. We see that the transportation cost can be minimized by the application of CRT under BCM.

Keywords: best candidate method, centroid ranking technique, fuzzy transportation problem, robust ranking technique, transportation problem

Procedia PDF Downloads 263

Authors: Doru Anastasiu Popescu, Dan Rădulescu

Abstract:

In this paper, we determine the similarity of two HTML web applications. We are going to use a genetic algorithm in order to determine the most significant web pages of each application (we are not going to use every web page of a site). Using these significant web pages, we will find the similarity value between the two applications. The algorithm is going to be efficient because we are going to use a reduced number of web pages for comparisons but it will return an approximate value of the similarity. The binary trees are used to keep the tags from the significant pages. The algorithm was implemented in Java language.

Keywords: Tag, HTML, web page, genetic algorithm, similarity value, binary tree

Procedia PDF Downloads 334

Authors: Madiha Khan, Sidrah Ramzan, Seemab Khan, Shahzad Hassan, Kamran Saeed

Abstract:

Measuring semantic similarity between texts is calculating semantic relatedness between texts using various techniques. Our web application (Measuring Relatedness of Concepts-MRC) allows user to input two text corpuses and get semantic similarity percentage between both using WordNet. Our application goes through five stages for the computation of semantic relatedness. Those stages are: Preprocessing (extracts keywords from content), Feature Extraction (classification of words into Parts-of-Speech), Synonyms Extraction (retrieves synonyms against each keyword), Measuring Similarity (using keywords and synonyms, similarity is measured) and Visualization (graphical representation of similarity measure). Hence the user can measure similarity on basis of features as well. The end result is a percentage score and the word(s) which form the basis of similarity between both texts with use of different tools on same platform. In future work we look forward for a Web as a live corpus application that provides a simpler and user friendly tool to compare documents and extract useful information.

Keywords: Graphviz representation, semantic relatedness, similarity measurement, WordNet similarity

Procedia PDF Downloads 210

Authors: Adnan A. Y. Mustafa

Abstract:

In this paper we present a quick technique to measure the similarity between binary images. The technique is based on a probabilistic mapping approach and is fast because only a minute percentage of the image pixels need to be compared to measure the similarity, and not the whole image. We exploit the power of the Probabilistic Matching Model for Binary Images (PMMBI) to arrive at an estimate of the similarity. We show that the estimate is a good approximation of the actual value, and the quality of the estimate can be improved further with increased image mappings. Furthermore, the technique is image size invariant; the similarity between big images can be measured as fast as that for small images. Examples of trials conducted on real images are presented.

Keywords: big images, binary images, image matching, image similarity

Procedia PDF Downloads 168

Authors: Ma. Gracia Corazon Cayanan, Kai Yuen Cheong, Li Sha

Abstract:

Training a language model for a minority language has been a challenging task. The lack of available corpora to train and fine-tune state-of-the-art language models is still a challenge in the area of Natural Language Processing (NLP). Moreover, the need for high computational resources and bulk data limit the attainment of this task. In this paper, we presented the following contributions: (1) we introduce and used a translation pair set of Tagalog and English (TL-EN) in pre-training a language model to a minority language resource; (2) we fine-tuned and evaluated top-ranking and pre-trained semantic textual similarity binary task (STSB) models, to both TL-EN and STS dataset pairs. (3) then, we reduced the size of the model to offset the need for high computational resources. Based on our results, the models that were pre-trained to translation pairs and STS pairs can perform well for STSB task. Also, having it reduced to a smaller dimension has no negative effect on the performance but rather has a notable increase on the similarity scores. Moreover, models that were pre-trained to a similar dataset have a tremendous effect on the model’s performance scores.

Keywords: semantic matching, semantic textual similarity binary task, low resource minority language, fine-tuning, dimension reduction, transformer models

Procedia PDF Downloads 180

Authors: Yang Song, Yifan Guo, Edward F. Gehringer

Abstract:

Fifty years of research has found great potential for peer assessment as a pedagogical approach. With peer assessment, not only do students receive more copious assessments; they also learn to become assessors. In recent decades, more educational peer assessments have been facilitated by online systems. Those online systems are designed differently to suit different class settings and student groups, but they basically fall into two categories: rating-based and ranking-based. The rating-based systems ask assessors to rate the artifacts one by one following some review rubrics. The ranking-based systems allow assessors to review a set of artifacts and give a rank for each of them. Though there are different systems and a large number of users of each category, there is no comprehensive comparison on which design leads to higher reliability. In this paper, we designed algorithms to evaluate assessors' reliabilities based on their rating/ranking against the global ranks of the artifacts they have reviewed. These algorithms are suitable for data from both rating-based and ranking-based peer assessment systems. The experiments were done based on more than 15,000 peer assessments from multiple peer assessment systems. We found that the assessors in ranking-based peer assessments are at least 10% more reliable than the assessors in rating-based peer assessments. Further analysis also demonstrated that the assessors in ranking-based assessments tend to assess the more differentiable artifacts correctly, but there is no such pattern for rating-based assessors.

Keywords: peer assessment, peer rating, peer ranking, reliability

Procedia PDF Downloads 410

Authors: Gi Ryung Song, Kyoung Seok Kim

Abstract:

This study reviewed the literature that focused on similarity of various characteristics such as values, personality, or demographics between employee and other elements in its organization for example employee with leader, job, and organization. We divided a body of this study into two parts and organized and demonstrated recent studies in first part. Three issues appeared in this part, which are statistical ways of measuring similarity, supervisor-subordinate similarity, and person-organization fit with person-job fit. In the latter part, based on the three issues of recent studies, we suggested three propositions about points that the recent studies missed or the studies did not orient. First proposition argued about the direction of similarity, which could also be interpreted as there is causal relation between employee and its workplace environments. Second, we suggested a consideration of eliminating common variance buried in one’s characteristics or its profiles. Third proposition was about the similarity of extra role behavior between individual and organization, and we treated this organization’s level of extra role behavior as a kind of its culture. In doing so, similarity of individual’s extra role behavior and organization’s has the meaning that individual’s congruence against their organization culture.

Keywords: similarity, person-organization fit, supervisor-subordinate similarity, literature review

Procedia PDF Downloads 251

Authors: Fatimah Zawani Abdullah, Shereena Mohd Arif, Nurul Malim

Abstract:

The study of molecular similarity search in chemical database is increasingly widespread, especially in the area of drug discovery. Similarity search is an application in the field of Chemoinformatics to measure the similarity between the molecular structure which is known as the query and the structure of chemical compounds in the database. Similarity search is also one of the approaches in virtual screening which involves computational techniques and scoring the probabilities of activity. The main objective of this work is to determine the best fingerprint when compared to the other five fingerprints selected in this study using PubChem chemical dataset. This paper will discuss the similarity searching process conducted using 6 types of descriptors, which are ECFP4, ECFC4, FCFP4, FCFC4, SRECFC4 and SRFCFC4 on 15 activity classes of PubChem dataset using Tanimoto coefficient to calculate the similarity between the query structures and each of the database structure. The results suggest that ECFP4 performs the best to be used with Tanimoto coefficient in the PubChem dataset.

Keywords: 2D fingerprints, Tanimoto, PubChem, similarity searching, chemoinformatics

Procedia PDF Downloads 264

Authors: T. G. Asmitha Chandini

Abstract:

Web-based image search re-ranking, as an successful method to get better the results. In a query keyword, the first stair is store the images is first retrieve based on the text-based information. The user to select a query keywordimage, by using this query keyword other images are re-ranked based on their visual properties with images.Now a day to day, people projected to match images in a semantic space which is used attributes or reference classes closely related to the basis of semantic image. though, understanding a worldwide visual semantic space to demonstrate highly different images from the web is difficult and inefficient. The re-ranking images, which automatically offline part learns dissimilar semantic spaces for different query keywords. The features of images are projected into their related semantic spaces to get particular images. At the online stage, images are re-ranked by compare their semantic signatures obtained the semantic précised by the query keyword image. The query-specific semantic signatures extensively improve both the proper and efficiency of image re-ranking.

Keywords: Query, keyword, image, re-ranking, semantic, signature

Procedia PDF Downloads 529

Authors: E. Gouya Bozorg

Abstract:

The issue of ranking of sports teams, in particular soccer teams is of primary importance in the professional sports. However, it is still based on classical statistics and models outside of area of mathematics. Rigorous mathematics and then statistics despite the expectation held of them have not been able to effectively engage in the issue of ranking. It is something that requires serious pathology. The purpose of this study is to change the approach to get closer to mathematics proper for using in the ranking. We recommend using theoretical mathematics as a good option because it can hermeneutically obtain the theoretical concepts and criteria needful for the ranking from everyday language of a League. We have proposed a framework that puts the issue of ranking into a new space that we have applied in soccer as a case study. This is an experimental and theoretical study on the issue of ranking in a professional soccer league based on theoretical mathematics, followed by theoretical statistics. First, we showed the theoretical definition of constant number Є = 1.33 or ‘golden number’ of a soccer league. Then, we have defined the ‘efficiency of a team’ by this number and formula of μ = (Pts / (k.Є)) – 1, in which Pts is a point obtained by a team in k number of games played. Moreover, K.Є index has been used to show the theoretical median line in the league table and to compare top teams and bottom teams. Theoretical coefficient of σ= 1 / (1+ (Ptx / Ptxn)) has also been defined that in every match between the teams x, xn, with respect to the ability of a team and the points of both of them Ptx, Ptxn, and it gives a performance point resulting in a special ranking for the League. And it has been useful particularly in evaluating the performance of weaker teams. The current theory has been examined for the statistical data of 4 major European Leagues during the period of 1998-2014. Results of this study showed that the issue of ranking is dependent on appropriate theoretical indicators of a League. These indicators allowed us to find different forms of ranking of teams in a league including the ‘special table’ of a league. Furthermore, on this basis the issue of a record of team has been revised and amended. In addition, the theory of ranking can be used to compare and classify the different leagues and tournaments. Experimental results obtained from archival statistics of major professional leagues in the world in the past two decades have confirmed the theory. This topic introduces a new theory for ranking of a soccer league. Moreover, this theory can be used to compare different leagues and tournaments.

Keywords: efficiency of a team, ranking, special table, theoretical mathematic

Procedia PDF Downloads 397

Authors: A. Brunel, A. Kernevez, F. Leclere, J. Trenteseaux

Abstract:

Today, wine quality is only evaluated by wine experts with their own different personal tastes, even if they may agree on some common features. So producers do not have any unbiased way to independently assess the quality of their products. A tool is here proposed to evaluate wine quality by an objective ranking based upon the variables entering wine elaboration, and analysed through principal component analysis (PCA) method. Actual climatic data are compared by measuring the relative distance between each considered wine, out of which the general ranking is performed.

Keywords: wine, grape, weather conditions, rating, climate, principal component analysis, metric analysis

Procedia PDF Downloads 288

Authors: Rihab Ben Lamine, Raoudha Ben Jemaa, Ikram Amous Ben Amor

Abstract:

This paper proposes a graph planning technique for semantic adaptable Web Services composition. First, we use an ontology based context model for extending Web Services descriptions with information about the most suitable context for its use. Then, we transform the composition problem into a semantic context aware graph planning problem to build the optimal service composition based on user's context. The construction of the planning graph is based on semantic context aware Web Service discovery that allows for each step to add most suitable Web Services in terms of semantic compatibility between the services parameters and their context similarity with the user's context. In the backward search step, semantic and contextual similarity scores are used to find best composed Web Services list. Finally, in the ranking step, a score is calculated for each best solution and a set of ranked solutions is returned to the user.

Keywords: semantic web service, web service composition, adaptation, context, graph planning

Procedia PDF Downloads 492

Authors: M. Kamel El-Sayed

Abstract:

The process of determining the degree of membership for an element to an uncertain concept has been found in many ways, using equivalence and symmetry relations in information systems. In the case of similarity, these methods did not take into account the degree of symmetry between elements. In this paper, we use a new definition for finding the membership based on the degree of symmetry. We provide an example to clarify the suggested methods and compare it with previous methods. This method opens the door to more accurate decisions in information systems.

Keywords: information system, uncertain concept, membership function, similarity relation, degree of similarity

Procedia PDF Downloads 194

Authors: Aydin M. Torkabadi, Ehsan Pourjavad

Abstract:

A decade after the introduction of Lean in Saskatchewan’s public healthcare system, its effectiveness remains a controversial subject among health researchers, workers, managers, and politicians. Therefore, developing a framework to quantitatively assess the Lean achievements is significant. This study investigates the success of initiatives across Saskatchewan health regions by recognizing the Lean healthcare criteria, measuring the success levels, comparing the regions, and identifying the areas for improvements. This study proposes an integrated intelligent computing approach by applying Fuzzy Inference System (FIS) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). FIS is used as an efficient approach to assess the Lean healthcare criteria, and TOPSIS is applied for ranking the values in regards to the level of leanness. Due to the innate uncertainty in decision maker judgments on criteria, principals of the fuzzy theory are applied. Finally, FIS-TOPSIS was established as an efficient technique in determining the lean merit in healthcare systems.

Keywords: lean healthcare, intelligent computing, fuzzy inference system, healthcare evaluation, technique for order of preference by similarity to ideal solution, multi-criteria decision making, MCDM

Procedia PDF Downloads 136

Authors: Salima Kouici, Abdelkader Khelladi

Abstract:

In this work, we begin with the presentation of the Tθ family of usual similarity measures concerning multidimensional binary data. Subsequently, some properties of these measures are proposed. Finally, the impact of the use of different inter-elements measures on the results of the Agglomerative Hierarchical Clustering Methods is studied.

Keywords: binary data, similarity measure, Tθ measures, agglomerative hierarchical clustering

Procedia PDF Downloads 449

Authors: Milica Karadzic, Lidija Jevric, Sanja Podunavac-Kuzmanovic, Strahinja Kovacevic, Anamarija Mandic, Katarina Penov Gasi, Marija Sakac, Aleksandar Okljesa, Andrea Nikolic

Abstract:

The present paper deals with chromatographic lipophilicity prediction of newly synthesized steroid derivatives. The prediction was achieved using in silico generated molecular descriptors and quantitative structure-retention relationship (QSRR) methodology with the artificial neural networks (ANN) approach. Chromatographic lipophilicity of the investigated compounds was expressed as retention factor value logk. For QSRR modeling, a feedforward back-propagation ANN with gradient descent learning algorithm was applied. Using the novel sum of ranking differences (SRD) method generated ANN models were ranked. The aim was to distinguish the most consistent QSRR model that can be found, and similarity or dissimilarity between the models that could be noticed. In this study, SRD was performed with average values of retention factor value logk as reference values. An excellent correlation between experimentally observed retention factor value logk and values predicted by the ANN was obtained with a correlation coefficient higher than 0.9890. Statistical results show that the established ANN models can be applied for required purpose. This article is based upon work from COST Action (TD1305), supported by COST (European Cooperation in Science and Technology).

Keywords: artificial neural networks, liquid chromatography, molecular descriptors, steroids, sum of ranking differences

Procedia PDF Downloads 291

Authors: Abdelkrim Alfalah, Lahcen Ouarbya, John Howroyd

Abstract:

This paper investigates and compares the performance of four existing distances and similarity measures between partitions. The partition measures considered are Rand Index (RI), Adjusted Rand Index (ARI), Variation of Information (VI), and Normalised Variation of Information (NVI). This work investigates the ability of these partition measures to capture three predefined intuitions: the variation within randomly generated partitions, the sensitivity to small perturbations, and finally the independence from the dataset scale. It has been shown that the Adjusted Rand Index performed well overall, with regards to these three intuitions.

Keywords: clustering, comparing partitions, similarity measure, partition distance, partition metric, similarity between partitions, clustering comparison.

Procedia PDF Downloads 153

Authors: Revoti Prasad Bora, Nikita Katyal, Saurabh Yadav

Abstract:

Pre-processing plays an important role in various image processing applications. Most of the time due to the similar nature of images, a particular pre-processing or a set of pre-processing steps are sufficient to produce the desired results. However, in the education domain, there is a wide variety of images in various aspects like images with line-based diagrams, chemical formulas, mathematical equations, etc. Hence a single pre-processing or a set of pre-processing steps may not yield good results. Therefore, a Deep Learning based approach for dynamically selecting a relevant pre-processing technique for each image is proposed. The proposed method works as a classifier to detect hidden patterns in the images and predicts the relevant pre-processing technique needed for the image. This approach experimented for an image similarity matching problem but it can be adapted to other use cases too. Experimental results showed significant improvement in average similarity ranking with the proposed method as opposed to static pre-processing techniques.

Keywords: deep-learning, classification, pre-processing, computer vision, image processing, educational data mining

Procedia PDF Downloads 121

Authors: K. P. Sandesh, M. H. Suman

Abstract:

Text processing plays an important role in information retrieval, data-mining, and web search. Measuring the similarity between the documents is an important operation in the text processing field. In this project, a new similarity measure is proposed. To compute the similarity between two documents with respect to a feature the proposed measure takes the following three cases into account: (1) The feature appears in both documents; (2) The feature appears in only one document and; (3) The feature appears in none of the documents. The proposed measure is extended to gauge the similarity between two sets of documents. The effectiveness of our measure is evaluated on several real-world data sets for text classification and clustering problems, especially in banking and health sectors. The results show that the performance obtained by the proposed measure is better than that achieved by the other measures.

Keywords: document classification, document clustering, entropy, accuracy, classifiers, clustering algorithms

Procedia PDF Downloads 485

Authors: Doru Anastasiu Popescu, Raducanu Dragos Ionut

Abstract:

In this paper the presentation of a tool which measures the similarity between two websites is made. The websites are compound only from webpages created with HTML. The tool uses three ways of calculating the similarity between two websites based on certain results already published. The first way compares all the webpages within a website, the second way compares a webpage with all the pages within the second website and the third way compares two webpages. Java programming language and technologies such as spring, Jsoup, log4j were used for the implementation of the tool.

Keywords: Java, Jsoup, HTM, spring

Procedia PDF Downloads 356

Authors: Deokho Kim, Wonwoo Lee, Jaewoong Lee, Teresa Ng, Gun-Ill Lee, Jiwon Jeong

Abstract:

This paper presents a method for improving object search accuracy using a deep learning model. A major limitation to provide accurate similarity with deep learning is the requirement of huge amount of data for training pairwise similarity scores (metrics), which is impractical to collect. Thus, similarity scores are usually trained with a relatively small dataset, which comes from a different domain, causing limited accuracy on measuring similarity. For this reason, this paper proposes a deep learning model that can be trained with a significantly small amount of data, a clustered data which of each cluster contains a set of visually similar images. In order to measure similarity distance with the proposed method, visual features of two images are extracted from intermediate layers of a convolutional neural network with various pooling methods, and the network is trained with pairwise similarity scores which is defined zero for images in identical cluster. The proposed method outperforms the state-of-the-art object similarity scoring techniques on evaluation for finding exact items. The proposed method achieves 86.5% of accuracy compared to the accuracy of the state-of-the-art technique, which is 59.9%. That is, an exact item can be found among four retrieved images with an accuracy of 86.5%, and the rest can possibly be similar products more than the accuracy. Therefore, the proposed method can greatly reduce the amount of training data with an order of magnitude as well as providing a reliable similarity metric.

Keywords: visual search, deep learning, convolutional neural network, machine learning

Procedia PDF Downloads 190

Authors: Ian A. McCulloh, Madelaine Zinser, Jesse Patsolic, Michael Ramos

Abstract:

Recommender systems are a common artificial intelligence (AI) application. For any given input, a search system will return a rank-ordered list of similar items. As users review returned items, they must decide when to halt the search and either revise search terms or conclude their requirement is novel with no similar items in the database. We present a statistically designed experiment that investigates the impact of similarity ratings on human judgement to conclude a search item is novel and halt the search. 450 participants were recruited from Amazon Mechanical Turk to render judgement across 12 decision tasks. We find the inclusion of ratings increases the human perception that items are novel. Percent similarity increases novelty discernment when compared with star-rated similarity or the absence of a rating. Ratings reduce the time to decide and improve decision confidence. This suggests the inclusion of similarity ratings can aid human decision-makers in knowledge search tasks.

Keywords: ratings, rankings, crowdsourcing, empirical studies, user studies, similarity measures, human-centered computing, novelty in information retrieval

Procedia PDF Downloads 92

Authors: Omid Shahmirzadi, Adam Lugowski, Kenneth Younge

Abstract:

Automatic measurement of semantic text similarity is an important task in natural language processing. In this paper, we evaluate the performance of different vector space models to perform this task. We address the real-world problem of modeling patent-to-patent similarity and compare TFIDF (and related extensions), topic models (e.g., latent semantic indexing), and neural models (e.g., paragraph vectors). Contrary to expectations, the added computational cost of text embedding methods is justified only when: 1) the target text is condensed; and 2) the similarity comparison is trivial. Otherwise, TFIDF performs surprisingly well in other cases: in particular for longer and more technical texts or for making finer-grained distinctions between nearest neighbors. Unexpectedly, extensions to the TFIDF method, such as adding noun phrases or calculating term weights incrementally, were not helpful in our context.

Keywords: big data, patent, text embedding, text similarity, vector space model

Procedia PDF Downloads 144