Jaro-Winkler Distance Research Articles

Identification of Synonyms Using Definition Similarities in Japanese Medical Device Adverse Event Terminology

Japanese medical device adverse events terminology, published by the Japan Federation of Medical Devices Associations (JFMDA terminology), contains entries for 89 terminology items, with each of the terminology entries created independently. It is necessary to establish and verify the consistency of these terminology entries and map them efficiently and accurately. Therefore, developing an automatic synonym detection tool is an important concern. Such tools for edit distances and distributed representations have achieved good performance in previous studies. The purpose of this study was to identify synonyms in JFMDA terminology and evaluate the accuracy using these algorithms. A total of 125 definition sentence pairs were created from the terminology as baselines. Edit distances (Levenshtein and Jaro–Winkler distance) and distributed representations (Word2vec, fastText, and Doc2vec) were employed for calculating similarities. Receiver operating characteristic analysis was carried out to evaluate the accuracy of synonym detection. A comparison of the accuracies of the algorithms showed that the Jaro–Winkler distance had the highest sensitivity, Doc2vec with DM had the highest specificity, and the Levenshtein distance had the highest value in area under the curve. Edit distances and Doc2vec makes it possible to obtain high accuracy in predicting synonyms in JFMDA terminology.

Spell corrector for Bangla language using Norvig’s Algorithm and Jaro-Winkler distance

In the Online World, especially in the social media platform most of us write without much regard to correct spelling and grammar. The spelling mistakes are much larger in proportion when it comes to Bangla language. In our paper, we presented a method for error detection and correction in Bangla words' spellings. Our system could detect a misspelled Bangla word and provide two following services- suggesting correct spellings for the word and correcting the word. We had used Norvig's Algorithm for the purpose but instead of using probabilities of the words to prepare the suggestions and corrections, we had used Jaro-Winkler distance. The previous works done in this field for Bangla language are either very slow or offers less accuracy. Our system successfully achieved a 97% accuracy when evaluated with 1000 Bangla words.

PENDETEKSIAN KEMIRIPAN TEKS DESKRIPSI DIRI PADA E-RECRUITMENT KARYAWAN DENGAN METODE RABIN KARP DAN JARO WINKLER DISTANCE

E-recruitment is a web-based application that is used to recruit employees for several available companies. The e-recruitment system has a plagiarism checker feature. Plagiarism itself has the meaning of actions that are considered fraud by taking someone's idea or writing without mentioning a reference and claiming it as their own. So that the e-recruitment application requires a plagiarism detector which functions to make it easier for companies to assess the characteristics of their prospective employees. The plagiarism checker feature applies a string matching algorithm to a text to search for common words between texts. There are several algorithms used for string matching, two of which are the Rabin-Karp and Jaro-Winkler algorithms. The Rabin-Karp algorithm is one of the algorithms that is suitable for solving multi-string pattern problems, while the Jaro Winkler Distance algorithm has advantages in terms of accuracy. An e-recruitment application with a plagiarism detection feature is developed and tested on various types of text, namely in the form of a self-description text of the prospective employee. From the experimental results, it was found that the two algorithms can be used to detect plagiarism in the applicant's self-description text, but it still has shortcomings in checking it.

Kombinasi Damerau Levenshtein dan Jaro-Winkler Distance Untuk Koreksi Kata Bahasa Inggris

Writing is one of the efforts made by the writer to express ideas and ideas to others. But sometimes when writing, there are many errors in typing spelling, especially English spelling, resulting in errors in capturing the meaning and meaning of the writing. To overcome this problem, we need a system that can detect word spelling errors. Damerau Levenshtein and Jaro Winkler Distance Algorithms are algorithms that can be used as a solution to detect English typing errors. From the test results, it can be concluded that the Damerau Levenshtein and Jaro-Winkler Distance are able to optimally detect word mismatches and look for similarities of words compared. The Damerau Levenshtein Distance works by finding the smallest distance value, while the Jaro-Winkler Distance works by finding the greatest proximity value of the string being compared. Using this algorithm, errors in writing the spelling of words can be minimized.
 
 Keywords— Algorithm; Damerau Levenshtein; Jaro Winkler; Spelling Cheker; String Matching.

Entity perception of Two-Step-Matching framework for public opinions

Entity perception of ambiguous user comments is a critical problem of target identification for huge amount of public opinions. In this paper, a Two-Step-Matching method is proposed to identify the precise target entity from multiple entities mentioned. Firstly, potential entities are extracted by BiLSTM-CRF model and characteristic words by TF-IDF model from public comments. Secondly, the first matching is implemented between potential entities and an official business directory by Jaro–Winkler distance algorithm. Then, in order to find the precise one, an industry-characteristic dictionary is developed into the second matching process. The precise entity is identified according to the count of characteristic words matching to industry-characteristic dictionary. In addition, associated rate (global indicator) and accuracy rate (sample indicator) are defined for evaluation of matching accuracy. The results for three data sets of public opinions about major public health events show that the highest associated rate and accuracy rate arrive at 0.93 and 0.95, averagely enhanced by 32% and 30% above the case of using the first matching process alone. This framework provides the method to find the true target entity of really wanted expression from public opinions.

Deteksi Plagiarisme Skripsi Mahasiswa dengan Metode Single-link Clustering dan Jaro-Winkler Distance

Abstract— The rise of plagiarism is one of the negative impacts of the development of information and communication technology. Plagiarism can occur anywhere. One of the examples is a university with the object of plagiarism as a student's final project. So we need a system to detect plagiarism so that it can suppress plagiarism in the college environment. In detecting the similarity of a writing will be faster if the writing has been grouped before compared to each other. Single-link clustering was chosen because it has a simple algorithm and can be implemented without the initial cluster. In plagiarism plagiarism usually changes the sentence structure so that it looks different Jaro-Winkler distance is chosen because it can detect similarities in paragraphs that have been changed in sentence structure because Jaro-Winkler distance has a flexible indexing with theoretical distance so that a word or character is considered the same. The stages in this study include data collection, preprocessing, grouping writing with Single-link clustering, comparing writing with jaro-winkler distance, and testing with precision and recall. After testing, the average value of precision was 84.37% and recall was 84.37% with a level of plagiarism of 99.1%. Keywords—: jaro-winkler distance; plagiarisme; single-link clustering.

JACOB Voice Chatbot Application Using Wit.ai for Providing Information in UMN

In this modern age, chatbots have been widely used as an information provider. The popularity of voice-based chatbots or voice chatbots has also increased with the presence of technologies such as Microsoft Cortana, Amazon Lex, Apple Siri, and Google Assistant. Interaction with voice chatbot is done using a voice which is more convenient than the text-based chatbots. Universitas Multimedia Nusantara (UMN) requires a tool to provide information for the Informatics joint-degree program. In this paper, a voice chatbot named JACOB is proposed. JACOB is built as a web-based application and implemented with Web Speech API. The sentences received by JACOB are translated using Wit.ai. The Jaro-Winkler distance is also applied in this research to reduce errors in identifying the user’s name. This implementation has been tested successfully, and the total satisfaction score measured using End User Computing Satisfaction is 78.53%.

Comparison of document similarity measurements in scientific writing using Jaro-Winkler Distance method and Paragraph Vector method

The purpose of this research is to study the methods of measuring the similarity of documents and tell us which is the most suitable for Indonesian Scientific Writing. This research method used was Jaro-Winkler Distance as method. Jaro-Winkler is a method that calculates the distance between strings and then measures the similarity. Doc2Vec (Paragraph Vector) is a method that aims to represent documents in vector form for comparison with the machine learning process. The results of this study compare the results of plagiarism detection between the Jaro-Winkler Distance method and the Doc2Vec method. The best measurement comparison method used is the accuracy of the comparison of documents and their speed. Using the dataset created, Doc2Vec outperformed the Jaro-Winkler Distance algorithm in comparing document similarities. Therefore, the development of a document similarity method will be easier in the future by using Doc2Vec (Paragraph Vector) in Indonesian scientific works.

Jaro–Winkler Distance Improvement For Approximate String Search Using Indexing Data For Multiuser Application

Word searching method has been developed in many ways and named as: Hamming Distance, Jaccard Distance, Jaro Distance, Jaro-Winkler Distance, Levenshtein Distance, etc. Those methods are used for lexicographic comparison to find words according to the similarity of the words which searched. The time needed for searching by using these words distance method can cause overhead as some difference user might try to search the same words all over. If these method is used in a multi user application where the user generally searching for some keywords repeatedly, then the user might have a longer searching time compared to exact search. In spite of this problem, we try to propose a method where the first search result of the previous user, will be recorded to the database for future usage by indexing the search keywords. In order to try this method, we use Jaro-winkler Distance method to search words. From the test result show that combining indexing and similarity word searching by using Jaro-Winkler Distance method can decrease the searching time to 90-92% compared to just using the Jaro-Winkler Distance method only. As the searched data increased, the processing time can be shorten.

Reusability ontology in business processes with similarity matching

The working technology will provide information and knowledge. Information and technology can be developed in various ways, by reusing the technologies. In this study modeled the ontology of SOPs using protege. Ontology will be matched between ontology A and B to obtain similarity and reuse ontology to create a more optimal ontology. Matching is a matching process between both ontologies to get the same value from both ontologies. Jaro-Winkler distance is used to find commonality between ontology. The result of the Jaro-Winkler distance has a value of 0 and 1, in matching will be obtained value close to 0 or 1. On matching ontology obtained two tests using 40% SPARQL query. In the test it uses Jaro-Winkler distance with a value of 0.67. This research yields matching value between ontology A and ontology B which is the same so that reuse ontology can be done for better ontology

Computer speech recognition to text for recite Holy Quran

Memorizing Holy Quran or Tahfidz is important to worship for Muslim around the world. This research proposed a solution in memorizing and learning Holy Quran easily. To help in remembering the sentence of Holy Quran, Fisher-Yates Shuffle had implemented for randomization of the letter of the Holy Quran. In this research, the sound of Holy Quran had recorded and it was converted into Arabic text to recognize the character of text. Jaro-Winkler was used for text matching algorithm, and Google Speech API help to define speech recognition. The result showed that Fisher-Yates Shuffle Algorithm was successfully applied in randomization with 15 times of experiments. And also, Jaro-Winkler Distance algorithm had performed well as text matching between text from speech recognition and Holy Quran text. The result showed that the percentage of accuracy was around 91% and an average of matching time was 1.9 ms.

Mendeteksi Plagiarism Pada Dokumen Proposal Skripsi Menggunakan Algoritma Jaro Winkler Distance

The practice of plagiarism among students in working on their thesis might happen. Plagiarism is done considering to the limited workmanship and lack of motivation to try with their own ability. It is necessary to provide a tool to prevent the act of plagiarism among the students. In this research, we built an application to detect the plagiarism on the thesis proposal. 
 The application applies the Jaro Winkler Distance algorithm to detect the similarity from several documents. The first phase to detect the similarity of the thesis proposals is The text preprocessing phase of the document such as case folding, tokenizing, stopwords removal and stemming. Precision and Recall formula will be used to analyze the performance of the method in application. 
 The test result of the data set shows the application could find the 80% relevant data that indicated the similarity. It means the application could contribute to detect the plagiarism on the thesis proposal document.

Improving name comparison similarity scores to reduce number of records for clerical review

IntroductionMany well established string comparators are currently used in data linkage. Jaro-Winkler distance is SA NT DataLink’s metric of choice for comparing personal names. However, due to Jaro-Winkler’s lower specificity we investigated if output scores could be transformed to produce scores more closely matching those assigned manually.
 Objectives and ApproachOur objective was to reduce the need for clerical review by modifying the Jaro-Winkler distance metric output scores. Clerical reviewers assigned similarity scores to pairs of first or last names from a database of approximately 2,000 random cases. By plotting the Jaro-Winkler scores against those assigned by the reviewers, a distinct radical function shape was observed. We then transformed the Jaro-Winkler scores by applying a power function where we gradually changed the exponent until we obtained the best fit with our clerically assigned scores. From the next linkage, two separate outputs were created (original and modified) and the results compared.
 ResultsTo assess the best fit we calculated the sum of squared errors for each of tested exponent values ranging from 1.1 to 6.0 (with 0.1 steps). The minimum sum of squared errors was achieved with exponent value of 4.6. We performed a probabilistic linkage for one decade of the Birth Registry records looking for familial links. Two separate linkage runs were conducted and clerically reviewed. In the second run, names were compared using the modified Jaro-Winkler comparator. This resulted in a reduced number of false positives. Though the lower-end threshold of the clerically reviewed “grey area” had to be lowered, the overall range was narrower resulting in less record pairs for clerical review.
 Conclusion/ImplicationsBy transforming the Jaro-Winkler scores, we reduced the number of records requiring clerical review. While only three linkage variables were affected, the resultant outcome was encouraging enough to consider exploring other possibilities for replicating clerical review knowledge in other comparators and metrics to reduce the demands for clerical review.

Novel Text Steganography Using Natural Language Processing and Part-of-Speech Tagging

The practice of transmitting secret data by using cover data is called steganography. At present, several versatile steganographic methods are available using different digital objects (e.g. image, audio, video, text, etc.) as cover to hide data. Accordingly, many methods of steganalysis have been explored to reveal statistical anomalies in stego object, through which presence of secret data can be detected. However, applications of steganography and steganalysis techniques are challenging when cover object is text, as text does not have any redundant bits. In this paper, an innovative text steganography approach is proposed which uses natural language text as cover as well as secret message. The concept of shared key is also used here, that holds the count of each parts-of-speech of secret message. This key is RSA encrypted and shared with communicative parties. Stego created by this method also is in natural language text. This method is successful as the stego keeps the original meaning of the text in gross which makes it robust and undetectable. It shows good result in capacity ratio; also, the similarity index has been assessed by Jaro–Winkler distance and Generalized Levenshtein distance.

Perbandingan Hasil Deteksi Plagiarisme Dokumen dengan Metode Jaro-Winkler Distance dan Metode Latent Semantic Analysis

Various methods are applied in the application of plagiarism detection to help check the similarity of a document. Jaro-Winkler Distance can measure the distance between two strings. However, this method basically depends on the position of the word. Latent Semantic Analysis emphasizes the words contained in the document regardless of its linguistic character. This study compares the results of plagiarism detection using the Jaro-Winkler Distance and the Latent Semantic Analysis method. From comparing results of Jaro-Winkler Distance method and Latent Semantic Analysis method, Jaro-Winkler Distance method is better than Latent Semantic Analysis method if using the same test data. Jaro-Winkler Distance method will give plagiarism result 100% and Latent Semantic Analysis method will give plagiarism result 97,14%.

Application of Pareto optimization method for ontology matching in nuclear reactor domain

Abstract This article describes the need for ontology matching and describes the methods to achieve the same. Efforts are put in the implementation of the semantic web based knowledge management system for nuclear domain which necessitated use of the methods for development of ontology matching. In order to exchange information in a distributed environment, ontology mapping has been used. The constraints in matching the ontology are also discussed. Pareto based ontology matching algorithm is used to find the similarity between two ontologies in the nuclear reactor domain. Algorithms like Jaro Winkler distance, Needleman Wunsch algorithm, Bigram, Kull Back and Cosine divergence are employed to demonstrate ontology matching. A case study was carried out to analysis the ontology matching in diversity in the nuclear reactor domain and same was illustrated.

Enhanced text mining approach based on ontology for clustering research project selection

Research project selection is an essential task for government and private agencies. When a huge number of research proposals are received, it is common to group them along with their similarities in research discipline areas. Then the grouped proposals are assigned to the appropriate experts for peer review. Existing approaches are not efficient to classify the document of project proposal correctly. Text-mining methods are used to solve the problem of classifying text documents automatically. In this paper, research proposals are classified based on the discipline areas and proposals in each discipline are grouped using the text-mining technique. Dice’s coefficient, Damerau–Levenshtein distance, Tversky index, Cosine similarity and Jaro–Winkler distance are used to find the similarity between the documents. The classification is used to predict the target class for each document proposal in the dataset accurately. In this work, Proposed Ensemble classifiers are used that contain various classifiers such as Kernel Support Vector Machine (KSVM), Self-Organizing Map (SOM), K-Nearest Neighbor (KNN) and Naive Bayes. These are the four classification methods used for proposed work and get an accuracy of results.

Linkage of health insurance data to the Netherlands Twin Register

ABSTRACTObjectivesWe compared four different approaches to record linkage, demonstrating its potential in enriching the Netherlands Twin Register (NTR) with healthcare data from a large health insurance provider (AHD), which covers approximately 26% of the population. Challenges included the fact that young twins share most of their linkage variables and overlapping identification numbers were absent. We extensively validated the linkage results and complemented them with an indicator of representativeness.ApproachSubjects born since 1986 were selected in the NTR (N=30,383) and AHD (N=1,532,675). Linkage variables were cleaned and harmonised and linkage keys were chosen that were strong enough to uniquely identify most subjects in either dataset. The four linkages were deterministic, probabilistic, probabilistic with Jaro-Winkler distance, and probabilistic using encrypted identifiers.Validation took place by reviewing linkage variables and information like family relations; linked record pairs were categorised into correct, possible or false links. Furthermore, the consistency of linked data was reviewed focusing on Attention Deficit Hyperactivity Disorder (ADHD), a prevalent condition among children. Finally, we quantified the impact of the linkage on representativeness of the target population.ResultsProbabilistic linkage with distance calculation linked 7944 NTR subjects to the AHD with 82% correct, 17% possible, and 1% false links. It encompassed nearly all links identified by any of the other methods, which were more conservative.Information about medication was available in the NTR dataset for 40% of the linked subjects. Validation of the information about ADHD inside linked records revealed few discrepancies between the two datasets. Linkage had a modest effect on the representativeness of the population. The current linkage has confirmed a number of known ADHD cases and identified 146 new subjects using methylphenidate and 101 receiving psychiatric care for ADHD, who were not previously recognized as ADHD cases within the NTR. ConclusionRecord linkage with sources of data such as a health insurance database can be an efficient way of data collection in cohort research. Our study shows that record linkage is also possible for twin pairs. Although a minority of all NTR subjects were covered by the AHD, the linkage resulted in an enrichment of the NTR dataset. Exact information about the dosage and frequency of drugs was obtained without contacting subjects with detailed questionnaires. With a total of almost eight thousand retrieved records, the size of the linked dataset is sufficiently large for epidemiological research of non-rare conditions.

Pattern graph tracking-based stock price prediction using big data

Stock price forecasting is the most difficult field owing to irregularities. However, because stock prices sometimes show similar patterns and are determined by a variety of factors, we propose determining similar patterns in historical stock data to achieve daily stock prices with high prediction accuracy and potential rules for selecting the main factors that significantly affect the price, while simultaneously considering all factors. This study is intended at suggesting a new complex methodology that finds the optimal historical dataset with similar patterns according to various algorithms for each stock item and provides a more accurate prediction of daily stock price. First, we use a Dynamic Time Warping algorithm to find patterns with the most similar situation adjacent to a current pattern. Second, we select the determinants most affected by the stock price using feature selection based on Stepwise Regression Analysis. Moreover, we generate an artificial neural network model with selected features as training data for predicting the best stock price. Finally, we use Jaro–Winkler distance with Symbolic Aggregate approXimation (SAX) as a prediction accuracy measure to verify the accuracy of our model.

Text Documents Plagiarism Detection using Rabin-Karp and Jaro-Winkler Distance Algorithms

Plagiarism is an act that is considered by the university as a fraud by taking someone ideas or writings without mentioning the references and claimed as his own. Plagiarism detection system is generally implement string matching algorithm in a text document to search for common words between documents. There are some algorithms used for string matching, two of them are Rabin-Karp and Jaro-Winkler Distance algorithms. Rabin-Karp algorithm is one of compatible algorithms to solve the problem of multiple string patterns, while, Jaro-Winkler Distance algorithm has advantages in terms of time. A plagiarism detection application is developed and tested on different types of documents, i.e. doc, docx, pdf and txt. From the experimental results, we obtained that both of these algorithms can be used to perform plagiarism detection of those documents, but in terms of their effectiveness, Rabin-Karp algorithm is much more effective and faster in the process of detecting the document with the size more than 1000 KB.