Content Filtering

The volume of healthcare information available on the internet has exploded in recent years. Nowadays, many online healthcare platforms provide patients with detailed information about doctors. However, one of the most important challenges of such platforms is the lack of personalized services for supporting patients in selecting the best-suited doctors. In particular, it becomes extremely time-consuming and difficult for patients to search through all the available doctors. Recommender systems provide a solution to this problem by helping patients gain access to accommodating personalized services, specifically, finding doctors who match their preferences and needs. This paper proposes a hybrid content-based multi-criteria collaborative filtering approach for helping patients find the best-suited doctors who meet their preferences accurately. The proposed approach exploits multi-criteria decision making, doctor reputation score, and content information of doctors in order to increase the quality of recommendations and reduce the influence of data sparsity. The experimental results based on a real-world healthcare multi-criteria (MC) rating dataset show that the proposed approach works effectively with regard to predictive accuracy and coverage under extreme levels of sparsity.

Microsoft's AI chatbot Tay represents a pivotal case in conversational AI development, illustrating the critical importance of architectural safeguards and ethical constraints in machine learning systems. This technical examination dissects the architectural design flaws, implementation vulnerabilities, data processing weaknesses, and training regime deficiencies that contributed to Tay's rapid behavioral degradation when exposed to adversarial inputs. By identifying specific technical shortcomings-from inadequate content filtering to excessive parameter sensitivity and problematic reinforcement learning configurations the article establishes a framework for understanding conversational AI failures and outlines necessary implementation requirements for creating responsible systems that maintain ethical boundaries while preserving adaptive learning capabilities.

Large Language Models (LLMs) represent a paradigm shift in artificial intelligence, with enormous potential across myriad domains. However, the development and implementation of these models are accompanied by an urgent need for ethical guidelines and content moderation to prevent misuse and the spread of harmful content. This article undertakes a critical analysis of the unintended yet profound consequences such_linguistic limitationsoften referred to as "child safety locks"-have on LLMs' cognitive scope, their ability to form a holistic representation of human culture and experience, and their potential for creativity and research. We argue that while moderation is necessary, an overly restrictive approach risks leading to knowledge gaps, cultural myopia, semantic flattening, and a general reduction in the models' analytical and creative capacities. The article concludes with an appeal for the development of more sophisticated, context-sensitive moderation strategies that seek to balance safety with intellectual integrity and versatility.

The constant growth of online real estate information has emphasized the need for the creation and improvement of intelligent recommendation systems to help mitigate the difficulties associated with user decision making. This systematic review, following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines and criteria, investigates current approaches and models used in real estate recommendation systems, with a focus on papers published in 2019 and 2024. The review identifies four main techniques: content-based filtering, collaborative filtering, knowledge-based systems, and hybrid approaches. Key findings indicate a preference for deep learning models, specifically convolutional neural network and long-short term memory (CNN-LSTM) architectures, and highlight the most used property characteristics: price, number of rooms, size, and location. The research addresses several important challenges, including the cold start problem, data sparsity, and the importance of adaptive learning in dynamic markets. Potential future research fields are outlined, with a focus on hybrid model architectures, attention mechanisms, and explainable artificial intelligence (AI). This review provides a comprehensive overview of the field, enabling scholars and practitioners to improve the accuracy and user experience of real estate recommendation systems.

Situs imdb.com adalah salah satu sumber data film terbesar dan paling terpercaya yang menyediakan informasi tentang film-film dari berbagai genre. Dari jutaan data film yang pada situs imdb.com, terdapat beberapa genre yang paling sering digunakan, diantaranya action, adventure, animation, comedy, dll. Penerapan data mining bertujuan mempermudah mencapai tujuan yaitu untuk menganalisa pemilihan genre film dengan menggunakan algoritma Apriori. Algoritma Apriori dipilih untuk menentukan aturan asosiasi untuk menganalisis genre film apa yang cocok untuk pengunjung website imdb.com berdasarkan genre film yang pernah ia kunjungi. Metode yang digunakan adalah Pendekatan dilamulai dengan pengumpulan data, Algoritma Apriori dan Tahapan data mining yang terdiri dari Pemanaman bisnis, data, persiapan data, pemodelan dan evaluasi. Dengan penelitian ini mendapatkan hasil membantu produsen dan pembuat film dalam membuat strategi pemasaran didunia perfilman dengan efektif dan efesien dalam tingkat keberhasilan pembuatan film dari beberapa genre yang telah ada.

The rapid proliferation of social media platforms has enabled the spread of harmful content by bad actors, including bots, trolls, and purveyors of misinformation. This article examines how Large Language Models (LLMs), such as GPT-4 and BERT, can be leveraged to detect and mitigate the impact of these bad actors. Through advanced natural language processing (NLP) capabilities, LLMs offer significant improvements over traditional moderation tools. This paper explores the advantages, challenges, and ethical considerations of using LLMs for social media moderation and presents case studies where these models have been implemented effectively

The increasing use of the internet for health information brings challenges due to the complexity and abundance of data, leading to information overload. This highlights the necessity of implementing recommender systems (RSs) within the healthcare domain, with the aim of facilitating more effective and precise healthcare-related decisions for both healthcare providers and users. Health recommendation systems can suggest suitable healthcare items or services based on users' health conditions and needs, including medications, diagnoses, hospitals, doctors, and healthcare services. Despite their potential benefits, RSs encounter significant limitations, including data sparsity, which can lead to recommendations that are unreliable and misleading. Considering the increasing significance of health recommendation systems and the challenge of sparse data, we propose an effective approach to improve precision and coverage in recommending healthcare items or services. This aims to assist users and healthcare practitioners in making informed decisions tailored to their unique needs and health conditions. Empirical testing on two healthcare rating datasets, including sparse datasets, illustrate that our proposed approach outperforms baseline recommendation methods. It excels in improving both the precision and coverage of health-related recommendations, demonstrating effective handling of extremely sparse datasets.

A tourism recommendation system is a crucial solution to help tourists discover more diverse tourism destinations. A content-based approach in a recommender system can be an effective way of recommending items because it looks at the user's preference histories. For a cold-start problem in the tourism domain, where rating data or past access may not be found, we can treat the user's past-travel-photos as the histories data. Besides, the use of photos as an input makes the user experience seamless and more effortless. The current development in Artificial Intelligence-based services enable the possibilities to implement such experience. This research developed a Deep Learning-based mobile tourism recommender system that gives recommendations on local tourism destinations based on the user's favorite traveling photos. To provide a recommendation, we use cosine similarity to measure the similarity score between one's pictures and tourism destination's galleries thro...

The Internet's rapid growth has resulted in a surge of information, posing challenges for users trying to stay updated. The World Wide Web, with its diverse websites updated at varying intervals, necessitates efficient categorization of news articles. As the number of online news sources continues to rise, categorization becomes crucial for users to find information easily. Given the overwhelming volume of digital media information, categorizing news using algorithms and assigning multiple tags based on their category becomes essential. In the modern digital era, users face difficulties filtering and accessing relevant content tailored to their preferences due to the abundance of information from various sources. This study employs Natural Language Processing (NLP) methods for automated content filtering of RSS feeds. The developed system extracts feeds via RSS, generates headings and summaries, and employs NLP for categorization. This research has broad implications, offering opportunities for improved information management, personalized recommendations, and informed decision-making across diverse domains.

Informasi yang sangat banyak jumlahnya mendorong kebutuhan akan adanya sistem rekomendasi. Sistem rekomendasi merupakan sebuah perangkat lunak yang dapat digunakan untuk menghasilkan rekomendasi item yang dapat digunakan oleh user. Permasalahan yang ada dalam pembangunan sistem ini diantaranya terletak pada ketepatan rekomendasi yang dihasilkan. Teknik pendekatan paling sukses untuk membangun sistem rekomendasi ini salah satunya adalah dengan menggunakan collaborative filtering. Dari hasil penelitian sebelumnya, association rule memiliki nilai performansi tertinggi dalam menentukan rekomendasi bila dibandingkan dengan algoritma matrix factorization dan algoritma item-based recommendation. Sedangkan teknik most-frequent item secara sederhana dapat digunakan untuk menghasilkan rekomendasi berdasarkan item yang paling sering digunakan oleh user lain (neighbor). Pada penelitian ini dibuat sebuah sistem rekomendasi yang membandingkan antara 2 metode yaitu association rule dan most-frequent item. Sistem rekomendasi ini diterapkan pada data film dan menghasilkan rekomendasi berupa data item yang dapat ditonton oleh user. Hasil dari penelitian yang telah dilakukan adalah bahwa nilai precision yang dihasilkan dari metode associaton rule lebih tinggi dari nilai precision pada most frequent item pada semua skenario pengujian yang dilakukan. Nilai precision dari association rule adalah 0,38 sedangkan nilai precision most-frequent item adalah 0,23.

Informasi yang sangat banyak jumlahnya mendorong kebutuhan akan adanya sistem rekomendasi. Sistem rekomendasi merupakan sebuah perangkat lunak yang dapat digunakan untuk menghasilkan rekomendasi item yang dapat digunakan oleh user. Permasalahan yang ada dalam pembangunan sistem ini diantaranya terletak pada ketepatan rekomendasi yang dihasilkan. Teknik pendekatan paling sukses untuk membangun sistem rekomendasi ini salah satunya adalah dengan menggunakan collaborative filtering. Dari hasil penelitian sebelumnya, association rule memiliki nilai performansi tertinggi dalam menentukan rekomendasi bila dibandingkan dengan algoritma matrix factorization dan algoritma item-based recommendation. Sedangkan teknik most-frequent item secara sederhana dapat digunakan untuk menghasilkan rekomendasi berdasarkan item yang paling sering digunakan oleh user lain (neighbor). Pada penelitian ini dibuat sebuah sistem rekomendasi yang membandingkan antara 2 metode yaitu association rule dan most-freque...

Blog provides a simple way for people to share personal experiences and ideas, and has already become an important tool for people to communicate with each other. Due to the vast amount of information on a particular blog, it is often time consuming for reviewing and finding the blog-article to suit the reader's mind. This paper proposes a personalised blog recommendation system that utilises text mining and various recommendation techniques. It aims at providing personalized blog article recommendations with high efficiency and effectiveness. This paper surveys the landscape of actual and possible hybrid recommender systems, and introduces a novel hybrid recommendation method that combines text mining, collaborative filtering, content-based and demographic-based recommendations to recommend blogs.

Information passing through internet is generally unrestricted and uncontrollable and a good web content filter acts very much like a sieve. This paper looks at how citizens' internet viewing can be controlled using content filters to prevent access to illegal sites and malicious contents in Nigeria. Primary data were obtained by administering 100 questionnaires. The data was analyzed by a software package called Statistical Package for Social Sciences (SPSS). The result of the study shows that 66.4% of the respondents agreed that the internet is been abused and the abuse can be controlled by the use of content filters. The PHP, MySQL and Apache were used to design a content filter program. It was recommended that a lot still need to be done by public organizations, academic institutions, government and its agencies especially the Economic and Financial Crime Commission (EFCC) in Nigeria to control the internet abuse by the under aged and criminals.

Map-based visualization of different kinds of information, which can be geo-coded to a particular location, becomes more and more popular, as it is very well accepted and understood by end-users. However, a simple map interface is not enough if we aim to provide information about objects coming from vast and dynamic information spaces such as the Web or social networks are. In this paper, we propose a novel method for intelligent visualization of generic objects within a map interface called IntelliView, based on an extensive evaluation and ranking of objects including both content and collaborativebased approaches. We describe the method implementation in a web-based application called Present, which is aimed at recycling items in social networks together with an experiment aimed at evaluation of proposed approach.

In this paper, we describe a simple approach to filter unwanted web pages, according to their content. The result of this work is a demo of an application that is usable in realtime filtering and in non-real-time indexing of any given web pages. We describe a proposed technique step by step, while discussing possible alternative ways for each part. In the end we discuss the overall quality and proposed next steps that could lead to a fully usable business application.

Seperti halnya film animasi jepang atau bisa juga di sebut dengan film anime yang kini mulai banyak digemari oleh semua kalangan tanpa pandang umur, status dan profesi. Pada film anime jepang ini memiliki memiliki beberapa genre dari action, comedy, drama, romance, adventure dll,  bisa diakses seperti pada media online seperti website yang menyajikan macam-macam genre anime salah satunya adalah myanimelist.net pada web ini memiliki 14.000 film anime berbagi genre. Salah satu cara untuk meningkatkan daya tarik film anime adalah dengan pemanfaatan data pelihan genre yang sering di tonton oleh para penikmat film anime. Dengan data tersebut kita busa menganalisa tipe genre apa yang paling banyak di tonton, serta kecenderungan pemilihan tipe genre alternatif yang disukai oleh para penikmat film anime. Sehingga dengan data-data tersebut pihak creator bisa menentukan strategi untuk realese tipe genre yang akan dibuat berikutnya. algoritma apriori bagus untuk digunakan untuk pembentukan ite...

"Tendo como ponto de apoio pesquisa levada a cabo pelo Centro de Ensino e Pesquisa em Inovação da Fundação Getúlio Vargas – CEPI, o livro principia com uma revisão da história recente do Direito Autoral no Brasil à luz do surgimento das novas tecnologias, que permitiram que uma incontável quantidade de informação e conteúdo fosse propagada de forma imediata, dificultando, como consequência, o controle do uso de obras protegidas nas plataformas da internet e eventuais reflexos na liberdade de expressão e na caracterização de possíveis ilicitudes."
(Trecho do Prefácio da Ministra Nancy Andrighi sobre o artigo que inicia a obra  Direito Autoral e Internet: Diagnósticos e Perspectivas do Debate Brasileiro)

PrefixSpan adalah salah satu metode penggalian pola sekuensial yang menggunakan pendekatan Pattern Growth serta melakukan proyeksi sufiks terhadap basis data sekuens, sedangkan AprioriAll adalah salah satu metode penggalian pola sekuensial yang menggunakan pendekatan Apriori serta melakukan pembangkitan dan pengujian sekuens kandidat. Dengan bekal basis data sekuens yang terdiri atas ID kastamer, ID transaksi, dan item yang dibeli, kedua algoritma tersebut dapat menghasilkan pola-pola sekuensial yang memenuhi dukungan minimum yang diinginkan. Secara teoretik kinerja PrefixSpan lebih unggul daripada AprioriAll, sehingga perlu dilakukan analisis terhadap kinerja dari kedua algoritma tersebut. Dalam penelitian ini dilakukan analisis kinerja algoritma PrefixSpan dan AprioriAll pada penggalian pola sekuensial. PrefixSpan akan disusun menggunakan teknik proyeksi basis data Pseudoprojection dan AprioriAll akan disusun sesuai empat fase prosedurnya, yaitu fase pengurutan, fase litemset, fase transformasi, dan fase sekuens. Kemudian kedua aplikasi ini akan diuji ketangguhannya dengan menggali pola sekuensial dari basis data sekuens sintetik berukuran besar. Berdasarkan uji coba dapat disimpulkan bahwa secara umum durasi eksekusi dan utilisasi memori PrefixSpan lebih kecil daripada AprioriAll dan secara grafis dapat dikatakan bahwa PrefixSpan lebih skalabel dan terpercaya daripada AprioriAll.

Higher penalties may apply, and higher damages may be awarded, for offences and infringements involving the conversion of material into digital or electronic form. Paul Haskell-Dowland for providing me useful technical feedback and pointing out the word salad I usually wrote at 4am. Patryk Szewczyk for hammering out of me my overuse of the word "this". I can't thank you enough for this. While Trish Williams wasn't my supervisor when the end arrived and she abandoned me for greener pastures, I do need to thank her for getting me moving again after I had well and truly stalled. If it wasn't for Trish getting the writing wheels turning and her hatred of long run-on sentences, I don't think this thesis would ever seen the light of day. Finally, I need to thank my wife, Brenda. For putting up with my om grumpy periods, irritation and completely random outbursts. It was all the thesis, honest! v

The expansion of Digital Television and the convergence between conventional broadcasting and television over IP contributed to the gradual increase of the number of available channels and on demand video content. Moreover, the dissemination of the use of mobile devices like laptops, smartphones and tablets on everyday activities resulted in a shift of the traditional television viewing paradigm from the couch to everywhere, anytime from any device. Although this new scenario enables a great improvement in viewing experiences, it also brings new challenges given the overload of information that the viewer faces. Recommendation systems stand out as a possible solution to help a watcher on the selection of the content that best fits his/her preferences. This paper describes a web based system that helps the user navigating on broadcasted and online television content by implementing recommendations based on collaborative and content based filtering. The algorithms developed estimate the similarity between items and users and predict the rating that a user would assign to a particular item (television program, movie, etc.). To enable interoperability between different systems, programs' characteristics (title, genre, actors, etc.) are stored according to the TV-Anytime standard. The set of recommendations produced are presented through a Web Application that allows the user to interact with the system based on the obtained recommendations.

O artigo procura entender como os Estados interferem na Internet. De início, a Internet parecia ser um local resguardado do poder estatal, no entanto, a “independência” da Internet logo foi vista como utopia e, então, tentou-se criar vertentes de sua regulação. Assim como a Internet, a linha prevalente hoje da regulação da Internet, proposta por Lessig, de controle do código-fonte das aplicações e serviços da Internet, não deixou de evoluir nas décadas seguintes ao seu surgimento. Os Estados passaram a pressionar os intermediários a seguirem suas determinações por meio de responsabilidades e isenções quando suas leis são seguidas. No entanto, vê-se um crescente protagonismo das plataformas na regulação da liberdade de expressão, um dos direitos mais básicos exercidos online. Conclui-se, com o artigo, que a Internet nos impõe um novo modelo de regulação que tanto se adapta e se altera rapidamente, quanto interfere na capacidade estatal de controlar comportamentos e garantir direitos ...

The volume of healthcare information available on the internet has exploded in recent years. Nowadays, many online healthcare platforms provide patients with detailed information about doctors. However, one of the most important challenges of such platforms is the lack of personalized services for supporting patients in selecting the best-suited doctors. In particular, it becomes extremely time-consuming and difficult for patients to search through all the available doctors. Recommender systems provide a solution to this problem by helping patients gain access to accommodating personalized services, specifically, finding doctors who match their preferences and needs. This paper proposes a hybrid content-based multi-criteria collaborative filtering approach for helping patients find the best-suited doctors who meet their preferences accurately. The proposed approach exploits multi-criteria decision making, doctor reputation score, and content information of doctors in order to increase the quality of recommendations and reduce the influence of data sparsity. The experimental results based on a real-world healthcare multi-criteria (MC) rating dataset show that the proposed approach works effectively with regard to predictive accuracy and coverage under extreme levels of sparsity.

Information passing through internet is generally unrestricted and uncontrollable and a good web content filter acts very much like a sieve. This paper looks at how citizens' internet viewing can be controlled using content filters to prevent access to illegal sites and malicious contents in Nigeria. Primary data were obtained by administering 100 questionnaires. The data was analyzed by a software package called Statistical Package for Social Sciences (SPSS). The result of the study shows that 66.4% of the respondents agreed that the internet is been abused and the abuse can be controlled by the use of content filters. The PHP, MySQL and Apache were used to design a content filter program. It was recommended that a lot still need to be done by public organizations, academic institutions, government and its agencies especially the Economic and Financial Crime Commission (EFCC) in Nigeria to control the internet abuse by the under aged and criminals.

Today online social media are most valuable and essential member of human life. People can online communicate with their family, society and friends to exchange several type of information including text, images, audio & video data. Therefore, there is need of control of user over the contents which are published on his wall. Hence , user is able to avoid undesirable content that is displayed on his wall. But, currently social networks provide this service in very small extent. To provide this service, we plan a user defined filtering rules and machine learning categorization technique in this paper. Filtering rules grant users to personalize the filtering norms that are employed to the contents which are published on their walls. Automatic categorization based on content of messages into proposed categories is possible through machine learning technique. Also, proposed system can restrict the undesired images that are published on online social network (OSN) user private space by u...

This paper proposes a system that implements a content-based message filtering service for Online Social Networks (OSNs). Our system allows OSN users to have a direct control on the messages that are posted on their walls. This is done through a rule-based system, that allows a user to customize the filtering criteria, which is to be applied to their walls, and a Machine Learning based classifier which can automatically produce membership labels for the support of our content-based filtering mechanism. Keywords— On-line Social Networks; Information Filtering; Short Text Classification; Natural Language

Promoting recommender systems in real-world applications requires deep investigations with emphasis on their next generation. This survey offers a comprehensive and systematic review on recommender system development life cycles to enlighten researchers and practitioners. The paper conducts statistical research on published recommender systems indexed by Web of Science to get an overview of the state of the art. Based on the reviewed findings, we introduce taxonomies driven by the following five phases: initiation (architecture and data acquisition techniques), design (design types and techniques), development (implementation methods and algorithms), evaluation (metrics and measurement techniques) and application (domains of applications). A layered framework of recommender systems containing market strategy, data, recommender core, interaction, security and evaluation is proposed. Based on the framework, the existing advanced humanized techniques emerged from computational intelligence and some inspiring insights from computational economics and machine learning are provided for researchers to expand the novel aspects of recommender systems.

Seperti halnya film animasi jepang atau bisa juga di sebut dengan film anime yang kini mulai banyak digemari oleh semua kalangan tanpa pandang umur, status dan profesi. Pada film anime jepang ini memiliki memiliki beberapa genre dari action, comedy, drama, romance, adventure dll,  bisa diakses seperti pada media online seperti website yang menyajikan macam-macam genre anime salah satunya adalah myanimelist.net pada web ini memiliki 14.000 film anime berbagi genre. Salah satu cara untuk meningkatkan daya tarik film anime adalah dengan pemanfaatan data pelihan genre yang sering di tonton oleh para penikmat film anime. Dengan data tersebut kita busa menganalisa tipe genre apa yang paling banyak di tonton, serta kecenderungan pemilihan tipe genre alternatif yang disukai oleh para penikmat film anime. Sehingga dengan data-data tersebut pihak creator bisa menentukan strategi untuk realese tipe genre yang akan dibuat berikutnya. algoritma apriori bagus untuk digunakan untuk pembentukan ite...

A Recommendation System is a system that is used to recommend any product based on the user's rating. In the Recommendation System, we are searching for any product based on the reviews, if any product shows the best ratings, then we can select that product. Recommendation System has been used in a wide range of applications like Amazon, Netflix, Flipkart, Facebook, etc. E-Learning System is a system which helps to teach all the students in virtual way. Computers or Laptops and with Internet are used for working an E-Learning. E-Learning becomes a part of the Recommendation system. In this system, they are recommending those websites which is best to teach you, live doubt session like Coding Ninjas,

Este artigo, visa analisar o formato e a atuação dos provedores de aplicações, em especial buscadores e redes sociais, e como procuram justificar ações editoriais sobre o conteúdo de terceiros que circula em suas redes. Examinaremos brevemente a história jurídica que permitiu que os intermediários atuassem dessa forma. Por fim, analisaremos algumas possíveis soluções para o problema, tanto no âmbito norte-americano como no brasileiro. Conclui-se não haver, prime facie, ilegalidade na ação moderadora dos novos intermediários. Há liberdade de negócio e livre iniciativa das empresas. Mas isso não implica a necessidade de vácuo jurídico de normas que regulem os limites da atividade editorial desses intermediários. Principalmente quando se trata de plataformas que dominam amplamente o mercado em sua área de atuação.

Today's Online Social Networks are to communicate with people and share information with them. In this kind of network process user space is a space where scraps or temporal notes of its users can be displayed. But users have very less control on scraps posted on their space. Abused Scraps which are sending and received on the user private space are riddled by allowing OSN users to have a direct control on the scraps posted on their spaces through Filtering Rules and Context Based Filtering Mechanism. A Flexible Rule System, that allows users to customize the filtering criteria to be applied to their space. The Abused Scraps from the user space can be riddled by Machine Learning based soft classifier. The user has a system with a sophisticated approach to decide which scraps should be inserted into a Blacklist.

Homoglyphs are pairs of visual representations of Unicode characters that look similar to the human eye. Identifying homoglyphs is extremely useful for building a strong defence mechanism against many phishing and spoofing attacks, ID imitation, profanity abusing, etc. Although there is a list of discovered homoglyphs published by Unicode consortium, regular expansion of Unicode character scripts necessitates a robust and reliable algorithm that is capable of identifying all possible new homoglyphs. In this article, we first show that shallow Convolutional Neural Networks are capable of identifying homoglyphs. We propose two variations, both of which obtain very high accuracy (99.44%) on our benchmark dataset. We also report that adoption of transfer learning allows for another model to achieve 100% recall on our dataset. We ensemble these three methods to obtain 99.72% accuracy on our independent test dataset. These results illustrate the superiority of our ensembled model in detec...

This paper presents a survey and an evaluation of some of the existing methods and tools used for Internet filtering. A proposal for a filtering program is being presented. This program overcomes the limitations of filtering techniques by its hierarchical organization which combines URL, word matching, and labeling approaches.

This paper presents a survey and an evaluation of some of the existing methods and tools used for Internet filtering. A proposal for a filtering program is being presented. This program overcomes the limitations of filtering techniques by its hierarchical organization which combines URL, word matching, and labeling approaches.

Homoglyphs are pairs of visual representations of Unicode characters that look similar to the human eye. Identifying homoglyphs is extremely useful for building a strong defence mechanism against many phishing and spoofing attacks, ID imitation, profanity abusing, etc. Although there is a list of discovered homoglyphs published by Unicode consortium, regular expansion of Unicode character scripts necessitates a robust and reliable algorithm that is capable of identifying all possible new homoglyphs. In this article, we first show that shallow Convolutional Neural Networks are capable of identifying homoglyphs. We propose two variations, both of which obtain very high accuracy (99.44%) on our benchmark dataset. We also report that adoption of transfer learning allows for another model to achieve 100% recall on our dataset. We ensemble these three methods to obtain 99.72% accuracy on our independent test dataset. These results illustrate the superiority of our ensembled model in detec...

Seperti halnya film animasi jepang atau bisa juga di sebut dengan film anime yang kini mulai banyak digemari oleh semua kalangan tanpa pandang umur, status dan profesi. Pada film anime jepang ini memiliki memiliki beberapa genre dari action, comedy, drama, romance, adventure dll,  bisa diakses seperti pada media online seperti website yang menyajikan macam-macam genre anime salah satunya adalah myanimelist.net pada web ini memiliki 14.000 film anime berbagi genre. Salah satu cara untuk meningkatkan daya tarik film anime adalah dengan pemanfaatan data pelihan genre yang sering di tonton oleh para penikmat film anime. Dengan data tersebut kita busa menganalisa tipe genre apa yang paling banyak di tonton, serta kecenderungan pemilihan tipe genre alternatif yang disukai oleh para penikmat film anime. Sehingga dengan data-data tersebut pihak creator bisa menentukan strategi untuk realese tipe genre yang akan dibuat berikutnya. algoritma apriori bagus untuk digunakan untuk pembentukan ite...

Purpose – This paper aims to consider selected results from the Arts and Humanities Research Council (AHRC)-funded “Managing Access to the internet in Public Libraries” (MAIPLE) project, from 2012-2014. MAIPLE has explored the ways in which public library services manage use of the internet connections that they provide for the public. This included the how public library services balance their legal obligations and the needs of their communities in a public space and the ethical dilemmas that arise. Design/methodology/approach – The researchers used a mixed-method approach involving a review of the literature, legal analysis, a questionnaire survey and case studies in five public library authorities. Findings – UK public library services use a range of methods to regulate internet access. The research also confirms previous findings that filtering software is an ubiquitous tool for controlling access to and protecting library users from “inappropriate”, illegal and harmful internet...

Meningkatnya persaingan dalam industri video game menuntut para pengembang untuk meningkatkan daya tarik video game agar dapat bertahan dan bersaing. Salah satu jenis video game yang persaingannya ketat saat ini adalah MOBA (Multiplayer Online Battle Arena) game yang merupakan permainan yang bersifat real time yang dimainkan dengan kerjasama tim. Salah video game MOBA yang sedang ikut bersaing dalam memperebutkan konsumen saat ini yaitu Mobile Legend. Salah satu cara untuk meningkatkan daya tarik game MOBA adalah dengan pemanfaatan data pemilihan tipe karakter yang dimainkan oleh para pemain video game. Dengan data tersebut kita bisa menganalisa tipe karakter apa yang paling banyak dimainkan oleh player. Sehingga pihak pengembang bias menentukan tipe karakter yang akan release berikutnya. Dengan Algoritma Apriori, pengolahan data dan informasi nantinya akan dapat membat developer untuk menentukan karakter apa yang harus dipersiapkan untuk release, perencanaan strategi release karakt...

AbstrakRecommendation system sering dibangun dengan memanfaatkan data peringkat item dan data identitas pengguna. Data peringkat item merupakan data yang langka pada sistem yang baru dibangun. Sedangkan, pemberian data identitas pada recommendation system dapat menimbulkan kekhawatiran penyalahgunaan data identitas.Hybrid recommendation system memanfaatkan algoritma penggalian frequent itemset dan perbandingan keyword dapat memberikan daftar rekomendasi tanpa menggunakan data identitas pengguna dan data peringkat item. Penggalian frequent itemset dilakukan menggunakan algoritma FP-Growth. Sedangkan perbandingan keyword dilakukan dengan menghitung similaritas antara dokumen dengan pendekatan cosine similarity.Hybrid recommendation system memanfaatkan kombinasi penggalian frequent itemset dan perbandingan keyword dapat menghasilkan rekomendasi tanpa menggunakan identitas pengguna dan data peringkat dengan penggunaan ambang batas berupa minimum similarity, minimum support, dan jumlah r...

The rapid development of Internet technologies in recent decades has imposed a heavy information burden on users. This has led to the popularity of recommender systems, which provide advice to users about items they may like to examine. Collaborative Filtering (CF) is the most promising technique in recommender systems, providing personalized recommendations to users based on their previously expressed preferences and those of other similar users. This paper introduces a CF framework based on Fuzzy Association Rules And Multiple-level Similarity (FARAMS). FARAMS extended existing techniques by using fuzzy association rule mining, and takes advantage of product similarities in taxonomies to address data sparseness and non-transitive associations. Experimental results show that FARAMS improves prediction quality, as compared to similar approaches.

Penggalian frequent closed itemset merupakan salah satu bagian penting dari penggalian kaidah assosiasi (Association rule) karena dapat secara unik menentukan himpunan semua frequent itemsets dan supportnya.Berbagai algoritma penggalian frequent closed itemset telah ditemukan, diantaranya adalah algoritma CHARM dan algoritma DCI_CLOSED. Algoritma CHARM menggunakan format data vertikal diffset dan metode subsumption check untuk melakukan pemeriksaan duplikasi. Metode ini tidak efisien karena memerlukan penyimpanan semua frequent closed itemsets sebelumnya. Algoritma DCI_CLOSED menggunakan format data vertikal bitvectors dan menggunakan metode order preserving untuk melakukan pengecekan duplikasi. Metode ini efisien karena tidak memerlukan penyimpanan frequent closed itemsets sebelumnya. Berdasarkan riset dan teori yang berkaitan dengan penggalian frequent closed itemsets, belum ada algoritma yang mengintegrasikan penggunaan format data vertikal diffset dan pengecekan duplikasi tanpa melakukan penyimpanan semua frequent closed itemsets sebelumnya. Sehingga ada peluang penelitian untuk merancang perbaikan algoritma CHARM yang lebih efisien penggunaan memorinya. Metodenya adalah menggabungkan subsumption check pada cabang yang sedang dienumerasi dan metode order preserving, sehingga tabel hash tidak menyimpan semua frequent closed itemsets sebelumnya. Hasil penelitian menunjukkan algoritma perbaikan CHARM lebih efisien penggunaan memorinya bila dibandingkan dengan algoritma CHARM untuk nilai minimum support yang semakin kecil.

In Online Social Network, information filtering can be used for different and more responsive functions. This is owing to the fact that in OSN there is the possibility of posting or commenting other posts on particular public or private regions called general walls. Information filtering can be used to give users the ability to control the messages written on their walls automatically, by filtering out unwanted messages. OSN provide very little support to prevent unwanted messages on user walls. For instance Facebook permits users only to state who is allowed to insert messages in their walls. (friends, friends of friends, group of friends). Though no content based partialities are preserved and therefore it is not possible to prevent undesired communications. For instance political or offensive ones, no matter of the user who post them. By using the concept of content based techniques and rule based approach the contents from text and image can be filtered and unwanted posts can be block.

In the past few years, with the proliferation of mobile devices people are experiencing frequent communication and information exchange. For instance, in the context of people’s visits, it is often the case that each person carries out a smart phone, to get information about nearby places. When one visits some location, an application will recommend useful information according to its current location, preferences and past visits. This paper introduces the concept of smart routing as a recommender system for tourists that takes into account the dynamics of their personal user profiles. The concept relies on the hybrid recommender system for map routing which can be used as the alternative to the current systems. This system implements hybridization of content and collaborative-filtering. The purpose of this paper is to build a system for web and mobile based applications with the idea of recommending parameters like food, restaurants, hospitals, ATMs etc. according to users’ current location, preferences and past visits. This approach has its roots in information filtering and retrieval research. The proposed system has also taken care of the cold start problem for new users.

Penambangan Aturan Asosiasi adalah area data mining yang fokus untuk memangkas kunci kandidat, untuk menemukan set item frekuensi. Algoritma apriori klasik bagaimana pun memiliki batasan dan karena itu, memberikan peluang untuk memodifikasi algoritma ini. Algoritma baru yang termodifikasi dengan ukuran set dan frekuensi ukuran set memperkenalkan cara yang lebih efisien dan efektif untuk meminimalkan kunci kandidat. Tulisan ini menjelaskan perbandingan tahapan antara algoritma apriori klasik dan algoritma apriori termodifikasi.

Com a proximidade da aprovação da Diretiva do Parlamento Europeu e do Conselho relativa aos Direitos de Autor no Mercado Único Digital e da sua implementação nos Estados-membros da União Europeia, é essencial uma análise do seu artigo 13, que inicialmente implementaria a obrigatoriedade de  filtros  de  conteúdo  que  automaticamente  impedissem  o  upload  de  conteúdo que desrespeitasse direitos autorais. O objetivo do presente artigo é  o  de  fazer  uma  análise  técnica  dos  diferentes  modelos  de  filtragem  de  conteúdo,  compreendendo  quais  tipos  de  conteúdo  se  encaixam  e  como  dentro  destas  tecnologias,  observando  suas  limitações  tanto  técnicas  quanto  às  limitações  da  sua  utilização  pelo  direito,  aproximando  os  conhecimentos práticos e o estado da arte desses algoritmos às discussões do direito. Para além da análise dos próprios filtros e do artigo proposto, busca-se  diagnosticar  o  impacto  que  a  aprovação  da  dita  Diretiva  terá  nos  ambientes  online  europeus  e  na  sociedade  da  informação  como  um  todo, focando na clara incompatibilidade da proposta com a liberdade de expressão, utilizando o impacto da aprovação da proposta nos memes e na liberdade artística como exemplos. De maneira geral, o artigo afirma que, por diversas incompatibilidades, sejam elas materiais, nas altas margens de erro encontradas nos filtros, sejam elas relacionadas à subjetividade inerente à  análise  dos  direitos  autorais,  nem  em  2018  e  nem  em  nenhum  futuro  próximo  previsível  será  possível  uma  automação  da  análise  dos  direitos  autorais sem enormes danos tanto para a própria matéria quanto para os direitos fundamentais e a sociedade como um todo.

O presente artigo tem como objetivo analisar o papel dos filtros tecnológicos como uma ferramenta para a proteção dos direitos autorais na Internet. Filtros estes que hoje aparentemente se apresentam como o modelo mais promissor na remoção de conteúdo que viola direitos autorais na rede, através do seu sistema que impede o upload de conteúdo mediante identificação e comparação de Digital fingerprints em sua base de dados. Ao mesmo tempo o artigo pretende traçar um panorama da forma como a legislação e jurisprudência vem tratando o tema de remoção de conteúdo por violação de direitos autorais no Brasil, para isso, será feito um comparativo entre a Lei n° 9.610/98 (Lei de Direitos Autorais), a Lei n° 12.965/14 (Marco Civil da Internet) e as decisões do Superior Tribunal de Justiça (STJ) e os seus respectivos entendimentos acerca do dever, ou ausência dele, referente ao monitoramento de conteúdo pelos provedores de aplicação. Para compreender a progressão do entendimento do STJ sobre o tema será feita uma breve análise de duas decisões da Corte que tratam sobre remoção de conteúdo, considerando um lapso temporal de antes e após a vigência do Marco Civil da Internet. Após o comparativo, serão abordadas formas de remoção de conteúdo que não estão englobadas pela tríade acima citada, diante disso, será analisado o papel do Notice and Take Down, como uma ferramenta extrajudicial, e a “importação” do seu uso para o Brasil, bem como a remoção de conteúdo por violação dos Termos de Uso de uma plataforma, e seus reflexos (chilling effects) na violação à liberdade de expressão dos usuários. Por fim, o artigo propõe a reflexão se diante de um cenário de ausência de legislação específica sobre o tema e de métodos alternativos que aparentemente não resolvem a questão, se é possível promover e tornar prático um caráter preventivo das plataformas que utilizam os filtros tecnológicos, em prol da proteção dos direitos autorais na Internet.

Human resources departments of organizations receive infinite resumes on regular basis, making it more tedious for the HR person to find the best job candidates. In fact, an employer receives an average of 144 resumes per job opening. According to market status, companies depend extensively on assessment tools to select the best suitable candidates for job positions. While most of the companies are turning to software tools that quickly identify the best candidates, the process of evaluation involves challenges in both the sequence of actions by HR and the quality of execution for each procedure. The question is whether the right candidates were chosen, as per the requirement of the jobs. The aim of this study is to provide an efficient resume extraction tool using keyword matching pattern algorithm. Single keyword pattern matching means locating all occurrences of the search pattern in the input text string. It occurs naturally as part of data processing, text editing, text retrieval, and so on. Many text editors and programming languages have facilities for matching strings. The core objective of this work is providing a simplest and well-organized resume automation tool. This tool is fast in data retrieval and therefore saves time. The working logic of this involves the combination of two keyword Pattern matching algorithm (HP&KR). Using the proposed algorithm, the data is processed to determine which applicants are the best fit for a particular job based on a powerful combination of skills, work experience, location etc. The algorithm devised in this research keeps the organizations needs as the prime aspect in evaluating the candidates.

SEMARANG 2018 BAB I PENDAHULUAN 1.1 LATAR BELAKANG Peningkatan jumlah dokumen ilmiah yang ada menimbulkan kebutuhan akan suatu sistem yang dapat memberikan rekomendasi dokumen ilmiah yang baik. Recommendation system merupakan model aplikasi yang dibangun dari hasil pengamatan terhadap keadaan dan keinginan pengguna. Sistem ini memanfaatkan opini pengguna terhadap suatu item dalam domain atau kategori tertentu. Karena itu sistem ini memerlukan model rekomendasi yang tepat agar apa yang direkomendasikan sesuai dengan keinginan pengguna, serta mempermudah pengguna mengambil keputusan yang tepat (McGinty dan Smyth, 2006).