In recent decades, multimedia has been increasingly gen- erated and used in various fields and applications such as, in healthcare, where numerical data (test results), images (X-rays, CT, MRI scans), time series (EEG or ECG), video (endoscopy), audio (recorded doctor’s voice), and textual data (test reports, doctor’s notes etc.), as illustrated by Fig. 1, are often generated when trying to make a diagnosis for a complex disease. It is more popular on social media where text, emoticons, images, video and audio talks, are also often uploaded and displayed to help enhance the meaning and understanding of a conversation or a concept. Then, integrating and/or fusing these multimedia data- sets together to obtain as much useful information as possible to improve machine learning, has become a challenging task [17, 20].
Different kinds of modules have been created: self-study units, an exercise server, a multi-mediadata base, didactical instructions and LiLi - a search engine for physics’ elearning material are offered on a virtual learnung environment (VLE). The modules and especially LiLi are presented here.
This research shows a new approach for Location Based Services (LBS) to provide GPS track log management through Open Source Software and Open Standards compliant prototype system. Recently, the performance of low cost GPS receivers has greatly improved due to better GPS chips, enhanced receiver functionality and SBAS (Satellite Based Augmentation System). However, GPS signal are degraded in “urban canyons” and under forest canopy nearby. This research shows the effective way to minimize the GPS signal errors through developed quality filtering by PDOP (Positional Dilution of Precision) and GPS positioning status, and road network based map-matching functionality. The prototype system also provides POI (Point of Interest) management functionality which can store POI in DBMS with multi-mediadata such as picture, movie, sound, and memo and can be synchronized with GPS track point automatically based on the common date and time attributes. Therefore, the user does not be required to carry special devices for field survey, for example, GPS enabled cell-phone or camera. This research aimed to achieve an effective framework for field survey in order to support the route and POI management using the developed GPS track log management functionality. The system also offers participatory framework for building POI with multiple users together through the Internet. Moreover, the research will scope to expand real-time data collection functionality for future ubiquitous environment. The research outcome could contribute to various location-based data collection subjects which include real-time targets.
The integrated essence of multi-media application system is the optimized comprehensive coordination design, namely a synthetically large-scale computer network system, and the systematical integration consists of computer software, hardware, operating system technology, database technology and network communication technology as well as the integration of selection and collocation for different manufacturers’ product, which ultimate objective is the optimization of integrated performance, the functional operation after resembling all components as well as ingredient, and a system of low cost, high efficiency, symmetric performance, augment ability as well as maintainability. The current research for multi-media technology still focuses on surmounting numerous critical technologies of computer processing multimedia, such as compression as well as the multi-media information decompression technology, the multi-media information network transmission technology, the multi-media technology standardization, the multi-media storage, input as well as output, system software and developing instruments as well as platforms. This performance provided an analysis for the solution of designing network as well as hardware in the synthetically multi-media integration and provided a particular design scheme.
This paper describes some recent results of our collaborative work on developing a speech recognition system for the auto- matic transcription or media archives from the British Broad- casting Corporation (BBC). The material includes a wide di- versity of shows with their associated metadata. The latter are highly diverse in terms of completeness, reliability and accu- racy. First, we investigate how to improve lightly supervised acoustic training, when timestamp information is inaccurate and when speech deviates significantly from the transcription, and how to perform evaluations when no reference transcripts are available. An automatic timestamp correction method as well as a word and segment level combination approaches between the lightly supervised transcripts and the original programme scripts are presented which yield improved metadata. Exper- imental results show that systems trained using the improved metadata consistently outperform those trained with only the original lightly supervised decoding hypotheses. Secondly, we show that the recognition task may benefit from systems trained on a combination of in-domain and out-of-domain data. Work- ing with tandem HMMs, we describe Multi-level Adaptive Net- works, a novel technique for incorporating information from out-of domain posterior features using deep neural network. We show that it provides a substantial reduction in WER over other systems including a PLP-based baseline, in-domain tandem fea- tures, and the best out-of-domain tandem features.
Rapid Eye Multi-Media DSP v9 features real-time recording and live local viewing. The system is designed to record, search and transmit video, audio, and data transactions while simultaneously providing users with both live and post event evaluation. Benefits of the system include a highly interactive, full local graphical user interface (LocalView) with vector driven pan/tilt/zoom mouse control. The Rapid Clip feature makes remote clip generation to a USB device or the internal DVD-RW fast and easy. By using a Linux operating system, Rapid Eye Multi-Media DSP v9 is extremely secure, stable and reliable with standard storage capacity from 250 GB to 1 TB, expandable up to 4 TB.
CCT ContactPro (CP) is the perfect solution for the agent desktop in an Avaya multi- media call center environment. ContactPro empowers agents to effi ciently serve cus- tomers in today‘s demanding world of multi-channel communication. The state-of- the-art technology solution helps leverage existing infrastructure while decreasing contact center costs and increasing customer satisfaction.
We believe social media is a promising plat- form for both studying SUD-related human be- haviors as well as engaging the public for sub- stance abuse prevention and screening. In this study, we have focused on four main tasks (1) employing unsupervised features learning to take advantage of a large amount of unsupervised so- cial mediadata (2) employing multi-view feature learning to combine heterogeneous user informa- tion such as “likes” and “status updates” to learn a comprehensive user representation (3) building SUD prediction models based on learned user fea- tures (4) employing correlation analysis to obtain human-interpretable results. Our investigation has not only produced models with the state-of-the-art prediction performance (e.g., for all three types of SUD, our models achieved over 80% prediction accuracy based on AUC), but also demonstrated the benefits of incorporating unsupervised hetero- geneous user data for SUD prediction.
We present MTL based deep learning approach for the problem of detecting user stance with re- spect to a particular topic: “Demonetisation”, on Twitter’s code-mixed Hindi-English data gener- ated by bilingual users. The machine transliter- ated and translated corpus is given to the model. We empirically demonstrated the effectiveness of the proposed architecture. The proposed approach of jointly training the main and the auxiliary task proved to be the best-performing model so far for the code-mixed data, indicating that it is a promis- ing new direction in the automated assessment of stance. An accuracy of 63.2% is achieved from our proposed deep learning model based on multi-task learning at detecting stance in code-mixed data which is an improvement of more than 4.5% over- all accuracy when compared with current bench- mark results.
The open-source SUMMA Platform is a highly scalable distributed architecture for monitoring a large number of media broad- casts in parallel, with a lag behind actual broadcast time of at most a few minutes. The Platform offers a fully automated me- dia ingestion pipeline capable of recording live broadcasts, detection and transcrip- tion of spoken content, translation of all text (original or transcribed) into English, recognition and linking of Named Enti- ties, topic detection, clustering and cross- lingual multi-document summarization of related media items, and last but not least, extraction and storage of factual claims in these news items. Browser-based graphical user interfaces provide humans with aggre- gated information as well as structured ac- cess to individual news items stored in the Platform’s database.
In the conclusion to the National Youth Service Corps (NYSC) Reproductive Health & HIV/AIDS Prevention Project Manual UNICEF/NYSC (2003), it was established that “AIDS is killing us in Africa”; that as a pandemic, it is “the world‟s most deadly undeclared war, and Africa has so far born its brunt” (p. 62). However, recent media campaigns in this regards as evidenced in the streams of exposures, are suggesting otherwise in the sense that there is a declared media war costing billions in whatever currency. Several media strategies and media vehicles have been evolved in this regards, which amplifies the position of Carol Bellamy, UNICEF Executive Director, when she echoed that “global success in combating HIV/AIDS must be measured by its impact on our children and young people” in the area of the quantum of information they need and at their disposal in fighting the scourge. Providing information in varied forms have proven a challenge that is being met especially in Nigeria by the efforts of donor agencies such as UNICEF, The Gates Foundation and countless other non-governmental organizations. Television and radio stations across the country beam signals depicting abstinences as the best approaches towards cure and prevention in much the same way as the print media carry, most times, free messages informing the public on how best to prevent and treat the menace. These, thus, establish that HIV/AIDS awareness is very present in Nigerian with a sero-prevalence rate strong among the youth between 15-29 years. Since global attention was drawn to the identification and spread of HIV/AIDS in 1983, the media has been awash with claims and counter-claims on the debilitating scourge. The focus of the media glitz has been on creating awareness, prevention and management of HIV/AIDS. Invariably, the world acknowledges the devastating effects of the HIV/AIDS pandemic. Consequently in Nigeria, governments and non-governmental organizations have come together – some for selfish [financial] ends and others for purely altruistic purposes to generate needed awareness and campaign against the spread of this global scourge through every available media.
To explore the dynamics and interplay of or- ganizational creativity and innovation in a case of simultaneous routine media content production and new product development, we utilize the con- cept and theories of ambidexterity. Ambidextrous organizations are able to combine exploitation – i.e., efficiency in operations and successful com- petition in current mature markets – and explora- tion, which means developing new products and services as well as flexibly responding to environ- mental changes (Lavie, Stettner & Tushman, 2010; O’Reilly & Tushman, 2013; Raisch & Birkinshaw, 2008; Raisch, Birkinshaw, Probst & Tushman, 2009; Tushman & O’Reilly, 1996). Specifically, we draw on the research on contextual ambidex- terity as “the capacity to simultaneously achieve alignment and adaptability at a business-unit lev- el” (Gibson & Birkinshaw, 2004, p. 209; see also Birkinshaw & Gibson, 2004). These dimensions of ambidexterity define the organizational settings of our empirical study, in which the members of the development team are simultaneously responsible for innovating and developing new media products tionship between ongoing operations and innova-
The best technology to do this is a satellite IP Data backbone infrastructure. Terrestrial networks are expensive fat pipe access points which are required for every cable headend or ISP. Satellite overlay provides a store and forward content distributed model which eliminates congestion caused by repeat transmission in the terrestrial backbone. Internet access costs can be significantly reduces using satellite for cached content and terrestrial for occasional real-time access.
This study examined the factors that advertising practitioners in Nigeria consider when selecting media for advertising campaigns. The survey research method was applied to collect data with the structured questionnaire as the research instrument. 120 copies of the questionnaire were administered on the respondents who were members of the Media Independents Association of Nigeria (MIPAN) sectoral group of the Nigerian advertising industry. The purposive sampling technique was employed for selecting respondents in order to eliminate waste. 110 copies of the questionnaire were properly completed and returned representing 91.6 percent response rate. Findings revealed that reach was the most important factor that influenced media selection among advertising practitioners in Nigeria as indicated by 40.5 percent of the respondents. This was followed by cost of media as well as available budget for advertising campaigns. Other factors include prestige, image, rating and share of media as well as circulation with respect to the print media. Radio emerged as the most preferred advertising medium as indicated by 37 percent of the respondents. It was followed by newspaper, television and online media respectively. The major conclusion drawn from this study is that traditional media of mass communication are still the preferred advertising media among advertising practitioners in Nigeria. It was recommended that private media owners should increase the reach of their media outlets in order to attract more advertising revenue. It was further recommended that future researchers should investigate how Nigerian advertising practitioners determine some of the variables utilized in media selection. These include reach, prestige, image, and share of advertising media. Furthermore, it was recommended that constant research be conducted to determine the actual reach of media outlets in Nigeria as opposed to unsubstantiated claims by media owners.
In the previous chapter the paper “Aurora: a data stream management system” (Abadi, et al., 2003) was mentioned. This paper proposes a new data stream management system called Aurora that aims to solve these problems. In order to achieve this Abadi, et al. designed a system that consists of a development GUI application, a server that streams the data and a performance monitor GUI application. Together this system allows users to manage their data streams by developing a so called flow network that defines what Aurora should do with their data. Users can then monitor their application and modify it at any time. One of Aurora’s novel features is their stream-oriented query operators that can be chained together using their development GUI application. Although this way of defining the flow of an application is intuitive it also requires a certain amount of data base management expertise to be able to combine them into a functional flow. Additional processing steps that a data stream operator would like to perform need to be scripted as separate applications that process Aurora’s output. Another novel feature is Aurora’s load shedding which improves the Quality of Service under heavy load by discarding a fraction of the data from the data stream. A user can set the Quality of Service specification to control how and when load shedding occurs. Although Aurora itself is too technical this project can benefit from its concepts. This project will use Aurora’s idea of dividing stream operators into boxes with settings and giving the user the ability to combine these boxes to manage the stream.
research. The use of entropy and distributions of traffic features has received a lot of attention in the research community. Disclosing malicious traffic for network security using entropy based approach and power law distribution is proposed. To calculate entropy feature considered is packet size. Malware, most commonly known as malicious data is prevalent, arising a number of critical threat issues. With the increasing volume of contents users share through social media, the user is going to share large amount of information. Using power law distribution malware is detected which the users share in social media by making a comparison with Shannon entropy technique .
Some scholars propose that the quality of journalism can be measured by the way in which a journalist goes about his or her job (Picard, 2000). The expenditure of time and resources and the dedication with which they gather information, attend staff meetings, discussions and training, obtain background material and knowledge and produce an end product are all activities believed to create work of higher quality. Some stress that quality in journalism is best seen in an end product that is accurate and truthful, delivered with credibility in a context encouraging democratic and civil activities to benefit society (Meyer, 2010). In his book The Vanishing Newspaper, Philip Meyer says that quality journalism can drive growth in circulation and profitability, if news organizations can deliver products that influence important aspects of community life. Journalism gains credibility and influence by being a trusted source of news, analysis and investigative reporting of public affairs. Meyer goes on to say that quality journalism involves influence in local communities, accuracy in reporting and readability. He also stresses the importance of editing. This echoes the principle of truth in journalism stated by Scheuer (2008), who writes that truth is a necessary precondition of quality journalism, and products resulting from good journalism should be able to answer far more than just “who, when, what and where”, but should also be able to put news into context by explaining “how” and “why”. Thus, in the following, the focus is on journalist’s motivation to adhere to professional values and their criteria identification of work practice quality and content quality. The results show journalists’ ability to sense quality journalism under challenge of convergence media environment and cross-media constraints and how they want to react to these factors in practice.
As several businesses have started to use internet in order to promote themselves internationally, they are also viewing the worth of better quality localization of website through method of adapting culturally and translating a website suitable for various cultures and languages. Nowadays ,social media marketing is being applied widely by marketers all across the globe. With around 600 million Facebook users and another 200 million twitter users not to state several business related LinkedIn and different local or small based competitors, the social media interaction sheer scale is phenomenal. Several businesses are using the platforms of social media for marketing, developing customer service and brand awareness, but only few are integrating the two localization areas and social media marketing (JustinParks, 2011). According to Walter (2013) culture is regarded as one of the biggest components of how people interact not just how they say something but how they select the components they use to acquire a message across. Most of the people related with social media and race or usage of internet become less distinct when they initiate regarding how similar socioeconomic level groups of varied races compare. King (2012) has mentioned one example of how cultural variations influence social media is to have a view at several platforms of social media. Although most of these familiar platforms are American they are also familiar outside of America. But if people listen closely they can view varied cultures using similar platform of social media variedly and they can hear friction of culture between varied cultures. This friction of culture is not only between nonnative English and English speakers ,but it is also amongst native English speakers from varied countries. Thus, cultural background always has an influence on people‟s interactions on social media. Thus, this study aimed to identify factors influencing social media marketing in two different culture contexts.