Keyword extraction aims to extract words that are able to represent the corpus meaning. Keyword extraction has a crucial role in information retrieval, recommendation systems and corpora classification. In Persian language, keyword extraction is known as hard task due to Persian’s inherent complication. In this research work, we aim to address keyword extraction with a combination of statistical and Machine Learning as a novel approach to this problem. First the required preprocessing is applied to the corpora. Then three statistical methods and Bayesian classifier was utilized to the corpora to extract the keywords pattern. Also, a post processing methods was used to decrease the number of True Positive outputs. It should be pointed out that the built model can extract up to 20 keywords and they will be compared with keywords in the corresponding corpus. The evaluation results indicate that the proposed method, could extract keywords from scientific corpora (Specifically Thesis and Dissertations) with a good accuracy.

Keywords can present the main concepts of the text without human intervention according to the model. Keywords are important vocabulary words that describe the text and play a very important role in accurate and fast understanding of the content. The purpose of extracting keywords is to identify the subject of the text and the main content of the text in the shortest time. Keyword extraction plays an important role in the fields of text summarization, document labeling, information retrieval, and subject extraction from text. For example, summarizing the contents of large texts into smaller texts is difficult, but having keywords in the text can make you aware of the topics in the text. Identifying keywords from the text with common methods is time-consuming and costly. Keyword extraction methods can be classified into two types with observer and without observer. In general, the process of extracting keywords can be explained in such a way that first the text is converted into smaller units called the word, then the redundant words are removed and the remaining words are weighted, then the keywords are selected from these words. Our proposed method in this paper for identifying keywords is a method with observer. In this paper, we first calculate the word correlation matrix per document using a feed forward neural network and Word2Vec algorithm. Then, using the correlation matrix and a limited initial list of keywords, we extract the closest words in terms of similarity in the form of the list of nearest neighbors. Next we sort the last list in descending format, and select different percentages of words from the beginning of the list, and repeat the process of learning the neural network 10 times for each percentage and creating a correlation matrix and extracting the list of closest neighbors. Finally, we calculate the average accuracy, recall, and F-measure. We continue to do this until we get the best results in the evaluation, the results show that for the largest selection of 40% of the words from the beginning of the list of closest neighbors, the acceptable results are obtained. The algorithm has been tested on corpus with 800 news items that have been manually extracted by keywords, and laboratory results show that the accuracy of the suggested method will be 78%.

Conventional word spotting systems determine hypothesized keywords and their confidence score using a speech recognizer. Acceptance or rejection of these keywords is intended based on comparison of their scores with a specific threshold. It has been proved that confidence score prepared by recognizer is highly dependent on sub-word structure of each keyword. So comparing assigned scores to keywords without considering their sub-word units could causes degradation in overall performance. In this paper a novel method for confidence score normalization is proposed which is based on sub-word units of each keyword and linear programming algorithm. In proposed method, a keyword-dependent correction term is added to the score of the keyword to maximize separation of confidence score histograms of true and false occurrences. Our results show a 2% improvement in FOM compared to baseline system. Also, choosing an appropriate feature vector has been discussed in this paper.

The current study attempts to explore the characteristics of author-assigned keywords in research articles as important constituents of targeted search in academic communities. To this end, the keywords of 200 research papers in the field of applied linguistics, in terms of domain, degree of specificity, and relation to the titles, were analyzed. To supplement the findings, the keyword choice strategy of a number of researchers with publishing experience in the field was also investigated. The analysis revealed a considerable rate of title-keywords match, especially with respect to field-specific keywords. This finding points to the importance of users’ field-specific background knowledge in locating relevant information on the web. The examination of authors’ viewpoints and strategies, on the other hand, helped to bring to light the complex and non-clichéd nature of keyword selection. The significance of authors’ diverging and converging attitudes and their implications for enhancing the success rate of keyword search are discussed.

Keywords and key phrases are subsets of most relevant words or phrases that summarize contents of a document while they play a critical role in information and document retrieval. Keyword extraction from scientific text is challenging and time-consuming due to the technical and multi-subject nature of the text, while the number of documents requiring keywords is increasing. There are various algorithms and methods developed for automatic keyword extraction. Rapid Automatic Keyword Extraction (RAKE) is a popular algorithm in this domain. RAKE’s decisions are based on the observation that keywords generally contain multiple words and they rarely include stopwords and words with minimum lexical meanings. Candidate keywords are a set of single-word or multi-word sequences selected based on the scores assigned to them by some scoring criteria in RAKE.In this research, a new modified version of RAKE algorithm is proposed in which candidate keyword scoring scheme is improved to increase precision and recall in the keyword extraction process. The proposed algorithm is to cover some of the main weaknesses of RAKE algorithm, especially in Persian scientific documents. To study the weaknesses of RAKE algorithm and evaluating the proposed modified version of RAKE, a set of metadata of Persian theses and dissertations are used. The result of test and evaluation of the proposed algorithm confirm improvement in precision, recall and F-measure.We study effectiveness of RAKE in extracting keywords from Persian texts. We find that RAKE algorithm often extracts long phrases with redundant words on Persian texts, leading to low accuracy. In this paper, we study sources of scoring inefficiency of RAKE algorithm and propose an improved version of RAKE algorithm with a novel scoring mechanism. Our scoring mechanism overcomes some of the weaknesses in RAKE’s original scoring for Persian texts and yields better results. Our evaluations on Persian corpus demonstrate that our improved RAKE algorithm outperforms original RAKE algorithm by extracting more accurate keyword. Our results show that improved RAKE achieves more than 20% higher precision and recall on average compared to original RAKE.

When the World Wide Web provides suitable methods for producing and publishing information to scientists, the Web has become a mediator to publishing information. This environment has been formed billions of web pages that each of them has a special title, special content, special address and special purpose. Search engines provide a variety of facilities limit search results to raise the possibility of relevance in the retrieval results. One of these facilities is the limitation of the keywords and search terms to the title or URL. It can increase the possibility of results relevance significantly. Search engines claim what are limited to title and URL is most relevant. This research tried to compare the results relevant between results limited in title and URL in agricultural and Humanities areas from their users sights also it notice to Comparison of the presence of keywords in the title and URL between two areas and the relationship between search query numbers and matching keywords in title and their URLs. For this purpose, the number of 30 students in each area whom were in MA process and in doing their thesis was chosen. There was a significant relevant of the results that they limited their information needs to title and URL. There was significantly relevance in URL results in agricultural area, but there was not any significant difference between title and URL results in the humanities. For comparing the number of keywords in title and URL in two areas, 30 keywords in each area were chosen. There was not any significantly difference between the number of keywords in the title and URL of websites in two areas. To show relationship between number of search keyword and the matching of title and URL 45 keywords in each area were chosen. They were divided to three parts (one keyword, two keywords and three keywords). It was determined that if search keyword was less, the amount of matching between title and URL was more and if the matching between title and URL was more, the rate of relevance of website was more. The research concluded that if users with specific names limit their search in the URL, they can get more related results. Also they can get better results by controlling of their search keywords. Specific search engines can also increase the weighting of keywords in the URLs and rank more relevant results to their users.

Considering the importance of using indexing tools such as thesauri and Medical Subject Headings (MeSH) in storage and retrieval of medical information, this study was performed with the aim to determine the compliance rate of keywords of articles published in an English-language journal Shahid Beheshti University of Medical Sciences, Journal of Paramedical Sciences, with the MeSH.
This was a cross-sectional and descriptive study using applied method. The data gathering tool was a researcher-made checklist. In total, 1143 keywords extracted from 269 articles published in 24 issues of the journal; those keywords were compared with MeSH, and were classified in three groups: exact match, partial match and no match.
During the investigated years, on average, 277 keywords (24.2%) were exact matches, 409 keywords (35.8%) were partial matches, and the 457 keywords (40.0%) were not match with the terms contained in the MeSH. Keywords in match or match with the entry term categories formed the largest portion of partial match keywords.
The results show that the authors of this journal do not have sufficient knowledge on correct keyword selection with use of the MeSH, and it is necessary for them to be trained in the correct way. In addition, the editorial board of this English-language Iranian journal in the paramedical field should request authors to adjust their keywords based on the MeSH during article selection.

Selection of index keywords through MeSH (Medical subject Headings) is one of the requirements set by international standards, in writing articles for medical journals. Therefore, the study has been performed with the aim of specifying and comparing level of compliance through MeSH-2014, between those keywords extracted from the abstracts of those articles published in Iranian and American dental journals, which have been indexed in PubMed.
Material and The study was an applied-descriptive one. In total, 1191 keywords extracted from 290 articles from amongst four selected Iranian journals, and 1207 keywords extracted from 203 articles from amongst 2 selected American journals have been compared with MeSH, and divided into three “exact match”, “relative match”, and “no match” groups. Data were analyzed, applying descriptive and inferential statistics (frequency percentage, and Chi-square).
There was no significant difference between ratios of the keywords with “exact match”, “relative match”, and “no match”, in selected Iranian and American journals. Most of the keywords in those articles published in Iranian journals were stemmed jointly from the “title and abstract”; however, in American journals, these keywords were mostly stemmed from the “text, references, or authors’ mentalities”. The superior journal from amongst reviewed journals was J Dent (Tehran), in terms of quality of the keywords.
As far as the authors’ knowledge for selection of keywords in both groups, and in terms of recognizing preferred terms existing in MeSH is concerned, both groups are far from desirable level. The issue requires more attention to be paid to the value of index words, indices, and the importance of indexing, by both groups of authors. Moreover, the matter should be considered by editorial boards of journals; so that, through more qualitative valuation of all aspects of the articles especially keywords, increase would be made in level of accessibility to the articles, and also citing them as an evidence.

Keyword extraction is a key step in document indexing. Keywords are semantic and content-based descriptors of a document, which can be used in document retrieval and representation. In databases containing scientific documents, such as Ganj in Irannian Research Institue for Information Science and Technology (IranDoc), it is even more critical to assign meaningful keywords for documents, since the documents are from different academic disciplines and contain technical terms.As the number of scientific documents grows exponentially, having an automatic and intelligent keyword extraction technique is getting more critical. There are various keyword extraction techniques that are either based on statistical features of the text or machine learning approaches, and sometimes a combination of both. In this research, we propose a new keyword extraction method for Persian scientific documents based on recommender systems and case-based reasoning. The proposed method is designed based on case-based reasoning in which the main assumption is that similar documents share similar keywords. There are two main steps in the proposed approach: first, similar documents to a given new document are retrieved based on TFIDF and word2vec model, second, the candidate keywords are extracted from retrieved documents and ranked based on a new scoring scheme, and a set of keyword are selected from the candidate keywords based on their score. The proposed method is tested and avaluated on a set of documents of Ganj database in three different subject areas (Art, Humanities and Engineering), based on precision, recall and expert panel

Abstract Abstract various definitions has represented for the theory that are accessible through bibliographic approach and content analysis. This difference is visible in the keywords of the definitions. In this paper, the definitions of the theory were extracted from the literature and main keywords, synonyms, and frequency of them were calculated by using content analysis. Content analysis of 125 definitions were identified 9 main keywords and 67 synonyms of keywords. The main keywords found in this content analysis such as concept, explanation, relation, phenomenon, series, hypothesis, systematic, prediction, and method. Finally, a revise definition of theory was represented, that illustrates its concept and the main keywords in the literature. In general, in the definition given in this paper, the theory is presented, the bibliographic and empirical approach is considered, and it refers to the aspects of description, explanation, prediction and control in a process and in a state of leveling in the definition of the theory with the use of vocabulary The key is attention.