Text Mining Algorithms Research Articles

Functional technology foresight. A novel methodology to identify emerging technologies

The speed and complexity of the technology evolution faced by modern societies need new approaches to the analysis and understanding of the world. Indeed, an exclusive focus on technological goals can miss to recognize all the stakeholders of a technology and address real user needs; moreover, on the one hand low signals are becoming more and more important in fast evolving markets, on the other hand the excess of hype, fashions, or vested interests sometimes deeply alter indicators. However, the so called Big Data promise to be a huge low cost set of valuable information, available and affordable to all (SMEs included). But, analyzing them is not trivial especially if we deal with academic papers and patents. To tackle these issues, the present paper proposes to apply a powerful methodological tool called Functional Analysis to the Technology Foresight process. Actually the rigorous study of the functions, that an artefact should perform to satisfy the user needs, provides a universal and thus unifying point of view, which is able to correlate the user perspective on the product with its technical features. Functional reasoning has been applied to (i) detect possible patterns of development, spotting missing elements and highlighting strengths as well as potential sources of failure; (ii) to enhance traditional bibliometric tools such as the analysis of S-curves and (iii), integrated with a natural language processing analysis toolchain, tailored for patent documents, to identify emerging technologies. The paper describes the functional approach to technology foresight activity, presents how to integrate it with text mining algorithms and experts’ domain knowledge, and finally discusses its benefits in the context of Technology Foresight also from an economic point of view, showing that oresight is affordable also for Small and Medium Enterprises.

텍스트 마이닝 알고리즘을 이용한 기상청 연구개발분야 과제의 추세 분석

이 연구에서는 비정형 자료 분석 기법 중 하나인 텍스트 마이닝 기법으로 기상청 연구개발분야 과제의 동향에 대하여 분석하였다. 이를 위하여 용어사전을 구축하고, 전처리를 하여 용어-문서 행렬을 만들었다. 이것을 이용해 연도별 용어 빈도수를 측정하고, 자주 나타나는 단어들에 대해서는 상대도수의 변화에 대해서 관찰하였다. 그리고 회귀 분석을 사용하여 증가추세와 감소추세를 가지는 용어들을 파악하였다. 이러한 분석으로 기상청 최근 연구개발 분야의 트렌드를 파악하였다. 이와 같은 연구는 향후 기상청 연구개발에 관한 기초 자료로 사용될 수 있으며, 연구개발의 방향성과 청사진을 제시하는데 이용될 수 있을 것이다. This paper considers the analysis of patterns in meteorological research and development using a text-mining algorithm as the method of analyzing unstructured data. To analyze text data, we define a list of terms related to meteorological research and development, construct times series of a term-document matrix through data preprocessing, and identify terms that have upward or downward patterns over time. The proposed methodology is applied to multi-year plans funded by Korea Meteorological Administration research and development programs from 2011 to 2015.

Role of Text Mining in Detection of Plagiarism in Arabic Texts: An Architectural Perspective

The aim of the study of to design and text mining tool for plagiarism detection in Arabic document. Plagiarism is an act or instance of using or closely imitating the language and thoughts of another author without authorization and the representation of that author's work as one's own, as by not crediting the original author (El-Matarawy <em>et al</em>., 2013). Plagiarism detecting in Arabic language documents are difficult because of the complex linguistic structure of this language. Text mining in data mining has a very good role in the natural language processing. In this study, we present plagiarism detection architecture for comparison of Arabic texts to identify similarities using text-mining methods. A new text-mining algorithm namely Text Mining Algorithm for Plagiarism Detection (TMA-PD) is proposed to generate tokens from the given Arabic document. A new framework, which combines the familiar text mining methods <em>Topic Tracking</em>, <em>Clustering</em> and <em>Concept Linkage</em>, is also proposed in this research study. This tool thus reduces the time in preliminary part in the detection of plagiarism. As the present text mining tools do not have the feature to process Arabic documents, a new add-on will be developed and integrated in it. Software agents are also used for better comparisons and to find out more texts that are similar. The performance of this tool will be evaluated on a large data set of Arabic texts.

Mining unstructured content for recommender systems: an ensemble approach

Recommendation of textual documents requires indexing mechanisms to extract structured metadata for attribute-aware recommender systems. Applying a variety of text mining algorithms has the advantage of capturing different aspects of unstructured content, resulting in richer descriptions. However, it is difficult to integrate them into a unique model so that these descriptions can efficiently improve recommendation accuracy. This article proposes a generic model based on ensemble learning that combines simple text mining methods in a post-processing approach. After executing each text mining technique, each set of metadata of a particular type is applied to the recommender module, which generates attribute-specific rankings. Then, the resulting recommendations are ensembled to generate a final personalized ranking to the user. We evaluated our ensemble technique with two attribute-aware collaborative recommenders (k-Nearest Neighbors and BPR-Mapping) and we demonstrate its generality by means of comparisons among different types of ensembles. We used two datasets from different domains, the first is from the Brazilian Embrapa Agency of Technology Information website, whose documents are written in Portuguese language, and the second is the HetRec MovieLens 2k, published by the GroupLens Research Group, whose movies' storylines are written in English. The experiments show that, particularly to the k-NN recommender, better accuracy can be obtained when multiple metadata types are combined. The proposed approach is extensible and flexible to new indexing and recommendation techniques.

A novel procedure on next generation sequencing data analysis using text mining algorithm

BackgroundNext-generation sequencing (NGS) technologies have provided researchers with vast possibilities in various biological and biomedical research areas. Efficient data mining strategies are in high demand for large scale comparative and evolutional studies to be performed on the large amounts of data derived from NGS projects. Topic modeling is an active research field in machine learning and has been mainly used as an analytical tool to structure large textual corpora for data mining.MethodsWe report a novel procedure to analyse NGS data using topic modeling. It consists of four major procedures: NGS data retrieval, preprocessing, topic modeling, and data mining using Latent Dirichlet Allocation (LDA) topic outputs. The NGS data set of the Salmonella enterica strains were used as a case study to show the workflow of this procedure. The perplexity measurement of the topic numbers and the convergence efficiencies of Gibbs sampling were calculated and discussed for achieving the best result from the proposed procedure.ResultsThe output topics by LDA algorithms could be treated as features of Salmonella strains to accurately describe the genetic diversity of fliC gene in various serotypes. The results of a two-way hierarchical clustering and data matrix analysis on LDA-derived matrices successfully classified Salmonella serotypes based on the NGS data. The implementation of topic modeling in NGS data analysis procedure provides a new way to elucidate genetic information from NGS data, and identify the gene-phenotype relationships and biomarkers, especially in the era of biological and medical big data.ConclusionThe implementation of topic modeling in NGS data analysis provides a new way to elucidate genetic information from NGS data, and identify the gene-phenotype relationships and biomarkers, especially in the era of biological and medical big data.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-016-1075-9) contains supplementary material, which is available to authorized users.

Erratum to: Extraction and visualization of industrial service portfolios by text mining of 10-K annual reports

As more and more manufacturing companies accumulate profits from service provision, the ability to monitor the adoption of the industrial services of other companies grows more important. The purpose of this paper is to propose a data-driven methodology for extraction of the industrial service portfolio from a company’s annual report. In this approach, form 10-K, a special format of annual report regulated by the Security Exchange Commission in United States is utilized as the data source. Because this document type contains rich information on a company’s operating segments, industrial service information is easily retrieved. Given the sheer volume of such documents, however, manual inspection is impractical. In order to resolve this issue, a text-mining algorithm is applied to automatically examine word-usage patterns and to identify the service portfolio. Then, the service portfolio’s relative position in the market is visualized on a positioning map. Due to the multi-dimensionality of the data, self-organizing map (SOM) is used as an alternative visualization scheme. SOM enables easy identification of the major service clusters as well as niche areas in the market; these, in turn, provide valuable information pertinent to service development planning. Also, and not least, policy makers can utilize our methodology to detect the servitization trends of various industries.

A Bayesian Sampling Method for Product Feature Extraction From Large-Scale Textual Data

The authors of this work propose an algorithm that determines optimal search keyword combinations for querying online product data sources in order to minimize identification errors during the product feature extraction process. Data-driven product design methodologies based on acquiring and mining online product-feature-related data are presented with two fundamental challenges: (1) determining optimal search keywords that result in relevant product related data being returned and (2) determining how many search keywords are sufficient to minimize identification errors during the product feature extraction process. These challenges exist because online data, which is primarily textual in nature, may violate several statistical assumptions relating to the independence and identical distribution of samples relating to a query. Existing design methodologies have predetermined search terms that are used to acquire textual data online, which makes the resulting data acquired, a function of the quality of the search term(s) themselves. Furthermore, the lack of independence and identical distribution of text data from online sources impacts the quality of the acquired data. For example, a designer may search for a product feature using the term “screen,” which may return relevant results such as “the screen size is just perfect,” but may also contain irrelevant noise such as “researchers should really screen for this type of error.” A text mining algorithm is introduced to determine the optimal terms without labeled training data that would maximize the veracity of the data acquired to make a valid conclusion. A case study involving real-world smartphones is used to validate the proposed methodology.

Computing similarity of text-based assembly processes for knowledge retrieval and reuse

The objective of this research is to use text similarity algorithms to enable retrieval and knowledge sharing of text-based assembly process plans. In distributed design-manufacturing enterprises, there exists a need to: (1) establish and use “best practice” assembly process descriptions to ensure process design consistency across manufacturing locations and (2) leverage and adapt existing process design knowledge across product lines. In this research, the need for better communication is addressed by design knowledge reuse. Specifically, previously-authored assembly processes are retrieved from a centralized repository with a text-based similarity and retrieval algorithm. The similarity of forty-five text-based assembly work instruction pairs (obtained from ten work instruction sets) is computed using four text mining algorithms: (1) Word Overlap, (2) Jaccard Score, (3) TF–IDF and (4) Latent Semantic Analysis. The similarity scores are used to compare and retrieve similar work instructions from a repository of existing assembly processes descriptions. A survey is conducted to develop a baseline quantification of assembly work instruction similarity. The scores from each text comparison method are compared to the scores from the survey. A statistical hypothesis test shows the Jaccard method mimics human interpretation of assembly work instruction similarity better than the three other text comparison methods. However, the Jaccard method is insensitive to synonymy and polysemy of words. The Latent Semantic Analysis method is relatively insensitive to synonymy and polysemy of words; and was found to have a difference of 0.1 with respect to survey data. This indicates that Latent Semantic Analysis can be used to retrieve assembly work instructions authored in free text. By doing so, engineers will be presented with similar variants of assembly work instructions have been authored. This will allow engineers to compare and assess the efficiency of their assembly process and gain insight into how other facilities are performing similar assembly operations.

Futuristic data-driven scenario building: Incorporating text mining and fuzzy association rule mining into fuzzy cognitive map

Fuzzy cognitive maps (FCMs) are one of the representative techniques in developing scenarios that include future concepts and issues, as well as their causal relationships. The technique, initially dependent on deductive modeling of expert knowledge, suffered from inherent limitations of scope and subjectivity; though this lack has been partially addressed by the recent emergence of inductive modeling, the fact that inductive modeling uses a retrospective, historical data that often misses trend-breaking developments. Addressing this issue, the paper suggests the utilization of futuristic data, a collection of future-oriented opinions extracted from online communities of large participation, in scenario building. Because futuristic data is both large in scope and prospective in nature, we believe a methodology based on this particular data set addresses problems of subjectivity and myopia suffered by the previous modeling techniques. To this end, text mining (TM) and latent semantic analysis (LSA) algorithm are applied to extract scenario concepts from futuristic data in textual documents; and fuzzy association rule mining (FARM) technique is utilized to identify their causal weights based on if-then rules. To illustrate the utility of proposed approach, a case of electric vehicle is conducted. The suggested approach can improve the effectiveness and efficiency of scanning knowledge for scenario development.

A Study on Application of Data Mining in Ayurinformatics

the traditional medicine system of ancient India, which nourishes an individual's physical strength and intelligence along with maintaining a wonderful balance with the environment. Deviant to its nature, it has now become an inactive magnificence. Neither the research methodology nor the text book has changed over the past 50 years and paving way for the modern medicine system to progress. This work gives a convincing solution for Ayurveda to regain its lost glory with the incorporation of Information Technology, Data mining, Statistics and Bioinformatics, thus forming a new venture named Ayurinformatics. To prove this, relevant information's were collected from various literatures of Ayurveda and data mining algorithms. The findings reached by the implementation of data mining algorithms like k-Means algorithm, Canny Edge detection algorithm, Decision support systems, etc., for finding a better research methodology to identify a better treatment method, has made a significant cutback in both time and money than modern medicine system does. This was apparent by the identification of using methods like k-means algorithm to cluster ayurvedic medicines, prediction methods for Bronchial Carcinoma- a type of cancer, Canny Edge Detection Algorithm for Medicinal Plants Disease Identification and AVD with Transitive text mining algorithm for extracting ayurvedic texts from biological databanks. This work strongly enforces that Ayurinformatics gives a new way of life to Ayurveda with its innovative approach to research methodology and better treatment identification methods.

Featured Article: Genotation: Actionable knowledge for the scientific reader.

We present an article viewer application that allows a scientific reader to easily discover and share knowledge by linking genomics-related concepts to knowledge of disparate biomedical databases. High-throughput data streams generated by technical advancements have contributed to scientific knowledge discovery at an unprecedented rate. Biomedical Informaticists have created a diverse set of databases to store and retrieve the discovered knowledge. The diversity and abundance of such resources present biomedical researchers a challenge with knowledge discovery. These challenges highlight a need for a better informatics solution. We use a text mining algorithm, Genomine, to identify gene symbols from the text of a journal article. The identified symbols are supplemented with information from the GenoDB knowledgebase. Self-updating GenoDB contains information from NCBI Gene, Clinvar, Medgen, dbSNP, KEGG, PharmGKB, Uniprot, and Hugo Gene databases. The journal viewer is a web application accessible via a web browser. The features described herein are accessible on www.genotation.org The Genomine algorithm identifies gene symbols with an accuracy shown by .65 F-Score. GenoDB currently contains information regarding 59,905 gene symbols, 5633 drug-gene relationships, 5981 gene-disease relationships, and 713 pathways. This application provides scientific readers with actionable knowledge related to concepts of a manuscript. The reader will be able to save and share supplements to be visualized in a graphical manner. This provides convenient access to details of complex biological phenomena, enabling biomedical researchers to generate novel hypothesis to further our knowledge in human health. This manuscript presents a novel application that integrates genomic, proteomic, and pharmacogenomic information to supplement content of a biomedical manuscript and enable readers to automatically discover actionable knowledge.

Multi-Objective Model Selection (MOMS)-based Semi-Supervised Framework for Sentiment Analysis

Sentiment analysis has emerged as an active research field due to the rapid growth of user-generated content on the Internet. This research area analyzes the opinions and attitudes of masses toward products, movies, topics, individuals, and services. Various machine learning and text mining algorithms have been used for sentiment analysis and classification. The recent research concludes that domain-specific lexicons perform significantly better as compared to domain-independent lexicons. The proposed research aims at improving the performance of general-purpose lexicons utilizing machine learning algorithms. A semi-supervised framework based on “MOMS” is introduced in order to determine the feature weight by incorporating SentiWordNet, a well-known general-purpose sentiment lexicon. The feature weights are learned by support vector machine, and the classification performance is enhanced by using Multi-Objective Model Selection procedure. Subjectivity criterion is used to select the desired features, and the effects of feature selection with respect to their part-of-speech information are studied comprehensively. Experimental evaluation is performed on seven different benchmark datasets which includes Large movie review dataset, Multi-domain sentiment dataset, and Cornell movie review dataset. The comparison of the proposed approach is performed with state-of-the-art techniques, lexicon-based approaches, and other methods for sentiment analysis. The proposed framework results in high performance when compared to other research in this field.

Review on Text Mining Algorithms

Review on Text Mining Algorithms

Analysis of IP-related court decisions using e-discovery - a case study of Apotex Pty. Ltd. v Sanofi-Aventis Australia Pty. Ltd. &amp; Ors

The High Court of Australia concluded that methods of medical treatment are patentable with the judgement cited by Apotex Pty. Ltd. v Sanofi-Aventis Australia Pty. Ltd. & Ors. This is the first occasion the High Court has considered whether methods for the medical treatment of the human body are patentable inventions. This research paper investigates the High Court decision using an IP knowledge management and e-discovery approach, integrating techniques of ontology construction, text, and data mining algorithms, to evaluate the logical High Court decision made without the need to manually read, organise, and interpret all of the associated IP laws, opinions, and related cases. The ontological schema of the knowledge domain is constructed, based on the key terms that frequently appear in the opinions written by the High Court justices. K-means cluster analysis is applied to verify the linkages between the justices and their reason-based opinion statements. The research results demonstrate the consistency of the High Court decision, which can be derived and analysed scientifically using modern integrated methods of computer knowledge-based e-discovery. The research provides an innovative approach in accelerating IP court decision research where time to market and legal changes to market access requires immediate strategic management decision making and international market planning.

Analysis of IP-related court decisions using e-discovery - a case study of Apotex Pty. Ltd. v Sanofi-Aventis Australia Pty. Ltd. &amp; Ors

The High Court of Australia concluded that methods of medical treatment are patentable with the judgement cited by Apotex Pty. Ltd. v Sanofi-Aventis Australia Pty. Ltd. & Ors. This is the first occasion the High Court has considered whether methods for the medical treatment of the human body are patentable inventions. This research paper investigates the High Court decision using an IP knowledge management and e-discovery approach, integrating techniques of ontology construction, text, and data mining algorithms, to evaluate the logical High Court decision made without the need to manually read, organise, and interpret all of the associated IP laws, opinions, and related cases. The ontological schema of the knowledge domain is constructed, based on the key terms that frequently appear in the opinions written by the High Court justices. K-means cluster analysis is applied to verify the linkages between the justices and their reason-based opinion statements. The research results demonstrate the consistency of the High Court decision, which can be derived and analysed scientifically using modern integrated methods of computer knowledge-based e-discovery. The research provides an innovative approach in accelerating IP court decision research where time to market and legal changes to market access requires immediate strategic management decision making and international market planning.

Gene-Disease Interaction Retrieval from Multiple Sources: A Network Based Method.

The number of gene-related databases has been growing largely along with the research on genes of bioinformatics. Those databases are filled with various gene functions, pathways, interactions, and so forth, while much biomedical knowledge about human diseases is stored as text in all kinds of literatures. Researchers have developed many methods to extract structured biomedical knowledge. Some study and improve text mining algorithms to achieve efficiency in order to cover as many data sources as possible, while some build open source database to accept individual submissions in order to achieve accuracy. This paper combines both efforts and biomedical ontologies to build an interaction network of multiple biomedical ontologies, which guarantees its robustness as well as its wide coverage of biomedical publications. Upon the network, we accomplish an algorithm which discovers paths between concept pairs and shows potential relations.

PaloPro: a platform for knowledge extraction from big social data and the news

PaloPro is a platform that aggregates textual content from social media and news sites in different languages, analyses them using a series of text mining algorithms and provides advanced analytics to journalists and social media marketers. The platform capitalises on the abundance of social media sources and the information they provide for persons, products and events. In order to handle huge amounts of multilingual data that are collected continuously, we have adopted language independent techniques at all levels and from an engineering point of view, we have designed a system that takes advantage of parallel distributed computing technologies and cloud infrastructure. Different systems handle data aggregation, data processing and knowledge extraction and others deal with the integration and visualisation of knowledge. In this paper, we focus on two important text mining tasks, named entity recognition from texts and sentiment analysis to extract the sentiment associated with the corresponding identified entities.

Text Clustering Algorithm Based on Random Cluster Core

Nowadays clustering has become a popular text mining algorithm, but the huge data can put forward higher requirements for the accuracy and performance of text mining. In view of the performance bottleneck of traditional text clustering algorithm, this paper proposes a text clustering algorithm with random features. This is a kind of clustering algorithm based on text density, at the same time using the neighboring heuristic rules, the concept of random cluster is introduced, which effectively reduces the complexity of the distance calculation.

A method for K-Means seeds generation applied to text mining

In this paper, a methodology is proposed in order to produce a set of seeds later used as a starting point to K-Means-type unsupervised classification algorithms for text mining. Our proposal involves using the eigenvectors obtained from principal component analysis to extract initial seeds, upon appropriate treatment for search of lightly overlapping clusters which are also clearly identified by keywords. This work is motivated by the interest of the authors in the problem of identification of topics and themes previously unknown in short texts. Therefore, in order to validate the goodness of this method, it was applied on a sample of labeled e-mails (NG20) representing a gold standard within the field of text mining. Specifically, some corpora referenced in the literature have been used, configured in accordance to a mix of topics contained in the sample. The proposed method improves on the results of other state-of-the-art methods to which it is compared.

A Novel Graph Decomposition Approach to the Automatic Processing of Poorly Formalized Data: Innovative Ideas – A Management Case Study

Introduction In most cases, automatic processing can only be conducted on a structured and normalized data set. If the data set is unstructured, then most of data processing algorithms will fail to provide adequate results. The selection of a formalized model for the data is a process that is conducted manually. The main reason behind using manual approach is that even the state-of-the-art text mining algorithms are incapable of processing complex semantics of non-formalized and unstructured textual data. On the other hand, important patterns can be extracted only from the structured data. There are varieties of methods that provide basic data cleansing and structuring services: clusterization algorithms, normalization techniques, data cleansing, etc. A general automatic formalized model construction can be realized for more complex cases. This becomes possible due to high abstraction power of graph theory. However, the graph theory is to be used carefully--trying to apply this theory without considering the research field features can lead to models that lack needed level of details for further processing. Common approaches to the processing of connected and poorly structured data are based on graphs theory. Poorly formalized and unstructured data is one of the largest segments in the data processing. Experts agree that almost 80 to 85 percent of business-relevant information originates in unstructured form. The manual processing of unstructured data is costly and time-consuming. To interpret unstructured information such techniques as natural language processing (NLP), data mining, text analytics are used. Further, the patterns found can be organized and structured using mathematical graphs. Different kinds of graphs can be used to represent different features of data. The field of innovation management was selected to apply the results of the research, because the development of this novel approach to data structuration is a part of work on universities' innovation life cycle model (ILCM) and innovation management system. The basic components of innovation management are ideas and innovations. For example, the idea can be represented by unstructured textual descriptions and a group of illustrations. It is important to classify the idea based on its description. The manual classification becomes difficult in case of large number of ideas. The idea can be provided in the form of terms-rich description. Such description contains specific key words. There are many techniques to identify such words and phrases, however these approaches are leaving aside the problems of textual structure and semantics. There are other attributes that must be used in order to structure ideas. Considering all these attributes, the process of ideas management becomes fairly complicated, thus automatic tools become highly important for efficient innovations management. Back in 1986, Andrew H. Van de Ven has outlined four central problems in innovation management (Van de Ven, 1986): managing attention; managing ideas into good currency; managing part-whole relationships; institutional leadership. According to Van de Ven's 2004 paper, these cornerstone problems remain intact and continue to heavily influence the performance of innovation management process (Van de Ven and Engleman, 2004). In our work we will concentrate on ideas and innovations automatic structuration tasks which are in the scope of outlined problem of managing ideas into good currency. One of the major parts of this problem is to correctly define the most promising and optimal (in respect to given resources) idea. One of the best ways to estimate an idea would be to compare it with implemented analogues. The estimation can become rather complicated when considering ideas interconnections and dependencies on each other or on specific innovation. In the second section of the paper we will provide a literature review of automatic and semi-automatic graph-based decision-support approaches dealing with the processing of poorly formalized data as well as a basic review of innovation management literature. …