Information Retreival

Recommendation systems are essential tools to overcome the choice overload problem by suggesting items of interest to users. However, they suffer from a major challenge which is the so-called cold-start problem. The cold-start problem typically happens when the system does not have any form of data on new users and on new items. In this chapter, we describe the cold start problem in recommendation systems. We mainly focus on Collaborative Filtering (CF) systems which are the most popular approaches to build recommender systems and have been successfully employed in many real-world applications. Moreover, we discuss multiple scenarios that cold-start may happen in these systems and explain different solutions for them.

Abstrak – Pesatnya pertumbuhan repositori data teks menyulitkan orang-orang untuk mengakses informasi yang dibutuhkan dengan cara yang efesien dan efektif. Permasalahan ini muncul sebagai dampak banyaknya informasi yang terintegrasi dalam teks yang kurang terstruktur. Akibatnya, komputer tidak dapat melakukan interpretasi dengan mudah. Dalam makalah ini, disajikan Ekstraksi data akademik pada text-mining dengan metoda sequence matching dalam peranannya untuk menjawab kebutuhan masa depan dalam hal penyediaan informasi yang bermanfaat dan mudah diakses oleh khalayak akademik dan umum. Sebagai contoh Aplikasi sistem informasi akademik, dibahas mengenai informasi akademik seperti krs online, SPMB online, KHS online, jurnal online, library online dan masih banyak lagi kegiatan akademik yang online. Proses ekstraksi data akademik pada text mining dengan menggunakan teknik SM untuk menemukan dan mengekstrak informasi dari database akademik dan seluruh layanan pada web tersebut. Dilakukan pengujian dengan parameter precision, recall serta f-measure untuk mengevaluasi kecocokan informasi yang dibutuhkan, Sehingga didapatkan performansi kinerja dari sisfokampus sesuai dengan relevansi informasi yang dibutuhkan. Penggunaan metoda tersebut mempercepat pencarian data dan layanan web terkait informasi akademik dan non akademik. Tantangan dari web database akademik mining adalah jumlah data atau volume data yang pertambahannya semakin lama semakin banyak untuk mempermudah pengaksesan dengan pengembalian data baik online maupun offline dari source text mining dari web data mining. Penelitian ini telah memberikan solusi bahwa dengan ekstrasi data akademik pafa text mining terintegrasi dengan metoda SM memperlancar proses pencarian dan pengaksesan informasi akademik baik yang tidak tersetruktur (informasi dari portal) bersifat portal, dan tersetruktur (Informasi dari aplikasi).
Kata Kunci: text-mining, ekstraksi informasi, Sequence Matching, ekstraksi web dat,  data mining,

Perkembangan internet yang pesat menyebabkan perkembangan volume textual yang luar biasa. Karenanya dibutuhkan tool dan metoda cerdas yang dapat mengakses konten sesuai kebutuhan. Salah satu metoda ekstraksi informasi yang berkembang adalah metoda yang menggunakan Sequence Matching (SM) dan Simple Tree Matching (STM) atau gabungan dari keduanya (hybrid) untuk memenuhi berbagai bentuk pola web data mining terutama web database mining akademik. Dalam tulisan ini dibahas kinerja kedua metoda hybrid (SM-STM) tersebut dalam mengekstraksi informasi pada halaman web database mining akademik. Hasil percobaan menunjukan dengan meningkatnya jumlah volume data seiring dengan pertambahan transaksi akademik dan mahasiswa baru sebagai sampel dalam pembelajaran diperoleh peningkatan kinerja hybrid. Kinerja hybrid terbaik yang merupakan salah satu representasi tag sequence yang sesuai untuk kedua teknik sequence matching dan tree matching. Tahap analisis dan pengujian memberikan hasil bahwa metode hybrid yang dibangun terbukti bisa mendapatkan konten baru pada halaman web database mining akademik meskipun pada beberapa dataset masih terdapat noise namun dibandingkan dengan metoda sebelumnya lebih baik.

La croissance rapide de la quantité d’information a motivé l’émergence de nombreux domaines infor- matiques, dont les systèmes de résumé automatique (RA). Ces systèmes sont utilisés pour extraire les informations les plus pertinentes de l’énorme quantité de données.
Dans ce mémoire, nous travaillons sur un système de RA qui se concentre sur les résumés des multiples articles scientifiques. Ces résumés permettent de surmonter la redondance des articles de la communauté scientifique et de fournir des synthèses de documents attachées au même sujet.
Nous utilisons dans notre système la maximisation des traits (MT) qui incarne une nouvelle approche langage-agnostique et non paramétrique de RA. Elle génère des résumés de communauté de documents multiples en utilisant les structures traditionnelles du RA et un modèle à base de graphes en utilisant l’algorithme de la propagation d’activation (PA).
En somme, ce mémoire a pour objectif de combiner l’approche de la MT avec les systèmes de RA et d’évaluer les résumés extractifs produits, que ce soient des résumés génériques ou des résumés se basant sur une requête.

Formulaic sequences (FSs) are among the most commonly discussed and well-documented effective factors in oral fluency both in L1 and L2. The present study aims to investigate the effect of teaching a set of 140 FSs on Iranian EFL learners' oral fluency. The relationship between the use of FSs and different measures of oral fluency is also studied empirically. Forty-eight intermediate EFL learners took part in the study. The participants were randomly assigned into two experimental groups and one control group. One of the experimental groups was taught the FSs with spaced retrievals. The other experimental group was also taught the FSs but with non-spaced retrievals. The control group was taught no FSs. The posttest was conducted one week after the treatment. Multivariate analysis of variance (MANOVA) and other statistical procedures were used for analyzing the data. The results indicated that spaced productive retrieval of the FSs after their explicit and holistic teaching, helped the learners to have longer mean length of runs. The present findings have important implications for solving students' problems in speaking the second language. Therefore, the significance of spaced retrieval of FSs in language teaching and learning should not be neglected.

Asked to edit and critically review the exchange between philosopher Rafael Capurro and Linda Treude on the concept of information, the below linked document presents the final version of the exchange, including my contributions to copy and content editing as well as bibliographic suggestions from my review. The final version of the document can be accessed at http://www.capurro.de/moraes.html

- Jared Bielby


"The term information is omnipresent in academic research as well as in everyday life. Buzz words such as information society, information age or information overload are floating around everywhere. But what do we mean exactly with 'information' in different contexts?

The philosopher Rafael Capurro dealt with this question in the late 1970's in his PhD thesis on the concept of information. Especially the philosophical community was suffering at that time from a lack of reflection thereupon. Capurro tried to fill this gap. The following remark by the physicist and philosopher Carl Friedrich von Weizsäcker (1912-2007) had a particular impact on him.

"Today we are getting accustomed [...] to understand information as something different from matter and consciousness. What we have discovered is an old truth in a new place. It is the Platonic eidos, the Aristotelian form, expressed in words such that also a human being in the 20th century can learn about them."

Weizsäcker’s step back to the history of ideas motivated Capurro also to deal with ethical issues of the information society particularly since the rise of the internet. The handling of digital information requires new cultural skills that are addressed with the term 'information literacy.' In the last ten years many projects in the Library and Information Science field (LIS) include or are even focused on information literacy (IL). In many of these projects there is a pre-understanding that IL has to do with the ability to find, filter, and process digital information. At the same time, there is a gap between this pre-understanding and an the ongoing academic discussion on the concept of information. The following dialogue deals particularly with this gap."

~ Linda Treude

Background and Aim: The object of the study was to Analyze IFLA Library Reference Model (LRM) Entities and Attributes for Iranian Traditional Music Resources, Case Study: Morq-e Sahar Song.
Method: The study inherits an applied content analysis method. All Entities and Attributes of  IFlA LRM Model based on two checklists include:  Final report of IFlA LRM on August 2017 and Transition Mappings User Tasks, Entities, Attributes, and Relationships in FRBR, FRAD, and FRSAD mapped to their equivalents in the IFLA Library Reference Model Analyzed for Morq-e Sahar Song data elements. The research sample size involves Fifty different performances of Morq-e sahar Song.
Results: The results show that all the eleven Entities and the Thirty -Seven Attributes of IFLA-LRM except for Cartographic scale attribute parallels in Morq-e sahar  song data elements.
Conclusion: As a first step,  the results of the study in identifying conveniences of this model for organizing traditional Iranian music resources, realizing IFlA LRM Model’s Entities and Attributes which is appropriate in Iranian traditional music organization is operational.  The results are also practical for experts in Information organization and designers of the Information retrieval systems.
Keywords: Iranian Music Resources, Music  Information Retrieval,  IFLA Library Reference Model(IFLA LRM), Traditional music Resources, Morq-e sahar Song

This paper describes an extension of our work presented in the robust English-to-French bilingual task of the CLEF 2007 workshop, a knowledge-light approach for query translation in Cross-Language Information Retrieval systems. Our work is based on the direct translation of character n-grams, avoiding the need for word normalization during indexing or translation, and also dealing with out-of-vocabulary words. Moreover, since such a solution does not rely on language-specific processing, it can be used with languages of very different nature even when linguistic information and resources are scarce or unavailable. The results obtained have been very positive, and support the findings from our previous English-to-Spanish experiments.

the purpose of this paper is to study the tasks of natural language processing (nlp) in music information retrieval. music computational research in the field suggests that the area of nlp is close to recent research in music. so we can profit from this similarity search nlp techniques applied in music. in this paper, the similarity between both the above described area. nlp levels listed here refer to relevant tasks in computational study music. nlp brief introduction about the history of music in the background makes clear the following different aspects of the research about the music, given its proximity to the nlp, is introduced. finally, recent research on music computational research compared with similar ideas in NLP and presented guidelines for music information retrieval.

Abbreviations are commonly found instances of synonymy in Biomedical journal papers. Information retrieval systems that index paragraphs rather than full-text articles are more susceptible to term variation of this kind, since abbreviations are typically only defined once at the beginning of the text. One solution to this problem is to expand the user query automatically with all possible abbreviation instances for each query term. In this paper, we compare the effectiveness of two abbreviation expansion techniques on the TREC 2006 Genomics Track queries and collection. Our results show that for highly ambiguous abbreviations the query collocation effect isn't strong enough to deter the retrieval of erroneous passages. We conclude that full-text abbreviation resolution prior to passage indexing is the most appropriate approach to this problem.

Formulaic sequences (FSs) are among the most commonly discussed and well-documented effective factors in oral fluency both in L1 and L2. The present study aims to investigate the effect of teaching a set of 140 FSs on Iranian EFL learners' oral fluency. The relationship between the use of FSs and different measures of oral fluency is also studied empirically. Forty-eight intermediate EFL learners took part in the study. The participants were randomly assigned into two experimental groups and one control group. One of the experimental groups was taught the FSs with spaced retrievals. The other experimental group was also taught the FSs but with non-spaced retrievals. The control group was taught no FSs. The posttest was conducted one week after the treatment. Multivariate analysis of variance (MANOVA) and other statistical procedures were used for analyzing the data. The results indicated that spaced productive retrieval of the FSs after their explicit and holistic teaching, help...

Inspectors need accurate and efficient methods for obtaining geometric information about facilities for construction and infrastructure management. Laser scanners can capture millions of 3D points, which have mm accuracy, in minutes, and hence can streamline extraction of geometric information. Currently, extraction of geometric information is done manually. This paper describes a case study, which shows that manual approach is inefficient and error-prone. Aiming at improving the efficiency and effectiveness of extracting survey goals, this paper discusses computer interpretable representations of surveying goals as a first step towards extraction of surveying goals. The described approach targets enabling surveyors to compose surveying goals as <facility parts, features, operations> tuples. Reasoning mechanisms can transform user-defined surveying goals described by these tuples into a sequence of measurement operations to generate results.