Multimedia data mining: state of the art and challenges

Advances in multimedia data acquisition and storage technology have led to the growth of very large multimedia databases. Analyzing this huge amount of multimedia data to discover useful knowledge is a challenging problem. This challenge has opened the opportunity for research in Multimedia Data Mining (MDM). Multimedia data mining can be defined as the process of finding interesting patterns from media data such as audio, video, image and text that are not ordinarily accessible by basic queries and associated results. The motivation for doing MDM is to use the discovered patterns to improve decision making. MDM has therefore attracted significant research efforts in developing methods and tools to organize, manage, search and perform domain specific tasks for data from domains such as surveillance, meetings, broadcast news, sports, archives, movies, medical data, as well as personal and online media collections. This paper presents a survey on the problems and solutions in Multimedia Data Mining, approached from the following angles: feature extraction, transformation and representation techniques, data mining techniques, and current multimedia data mining systems in various application domains. We discuss main aspects of feature extraction, transformation and representation techniques. These aspects are: level of feature extraction, feature fusion, features synchronization, feature correlation discovery and accurate representation of multimedia data. Comparison of MDM techniques with state of the art video processing, audio processing and image processing techniques is also provided. Similarly, we compare MDM techniques with the state of the art data mining techniques involving clustering, classification, sequence pattern mining, association rule mining and visualization. We review current multimedia data mining systems in detail, grouping them according to problem formulations and approaches. The review includes supervised and unsupervised discovery of events and actions from one or more continuous sequences. We also do a detailed analysis to understand what has been achieved and what are the remaining gaps where future research efforts could be focussed. We then conclude this survey with a look at open research directions.     

Temperature dependence of vickers microhardness and creep of InBi single crystals

Variation of Vickers microhardness of InBi single crystals with temperature has been studied. Loading time dependence of the microhardness at different temperatures has been used for the creep study in the temperature range 30° – 85°C. The activation energy for creep has been evaluated and the results are discussed.     

What kind of #conversation is Twitter? Mining #psycholinguistic cues for emergency coordination

The information overload created by social media messages in emergency situations challenges response organizations to find targeted content and users. We aim to select useful messages by detecting the presence of conversation as an indicator of coordinated citizen action. Using simple linguistic indicators drawn from conversation analysis in social science, we model the presence of coordination in the communication landscape of Twitter¹ using a corpus of 1.5 million tweets for various disaster and non-disaster events spanning different periods, lengths of time, and varied social significance. Within replies, retweets and tweets that mention other Twitter users, we found that domain-independent, linguistic cues distinguish likely conversation from non-conversation in this online form of mediated communication. We demonstrate that these likely conversation subsets potentially contain more information than non-conversation subsets, whether or not the tweets are replies, retweets, or mention other Twitter users, as long as they reflect conversational properties. From a practical perspective, we have developed a model for trimming the candidate tweet corpus to identify a much smaller subset of data for submission to deeper, domain-dependent semantic analyses for the identification of actionable information nuggets for coordinated emergency response.     

An automated approach to beat-by-beat QT-interval analysis

The present work attempts to resolve some of the problems encountered in QT interval analysis and to describe a new algorithm that uses wavelets to analyze the QT interval on a beat-by-beat basis. The QT interval is further segmented into its contributing factors to find the variability in the repolarization phase. The technique and the algorithm have been tested on control and disease data from the standard MIT ECG library to prove the reproducibility and accuracy of the system. The technique has been implemented on a PC DSP add-on card to do the analysis on a beat-by-beat basis     

Examination of the thermophysical bond of insert-molded PC/C fiber composites

The adhesive strength of a thermophysical bond between two polymers has been examined using fracture mechanics. Bimaterial composite specimens were constructed by injecting C fiber polyetheretherketone (PEEK) into a mold containing one-half of a polycarbonate (PC) dogbone. The resulting series specimens were notched at the interface and tested in tension. Adhesion of the two materials was reasonably good, as demonstrated by fracture surfaces that showed a mixture of PC and C fiber PEEK fragments. Interfacial fracture energy of the composite was approximately 1.5 kJ/m², which is comparable to the cohesive strength of amorphous commodity polymers. Variations in test speed (below the glass transition temperature of the two components, approximately 140°C) had no appreciable affect on stiffness or fracture energy. However, fracture energies decreased slightly as temperature increased.     

Optimisation of the alkaline peroxide pretreatment for the delignification of rice straw and its applications

Alkaline peroxide pretreatment for the delignification of rice straw was optimised by varying the concentrations of H₂O₂ and NaOH and changing the temperature and duration of the pretreatment. Changes in the lignin content, content of total carbohydrates and weight loss were measured during the pretreatments. Maximum delignification of 62% was obtained by pretreating rice straw at 50°C for 5h with 1.5% (w/v) NaOH and 1% (v/v) H₂O₂, The preferential loss of hemicellulose and lignin from the straw resulted in an increase in the cellulose content of the insoluble residue after pretreatment from 47% (untreated) to 67.79% (treated). The product of this treatment is characterised by having higher cellulose digestibility than untreated rice straw. It also has use as a carbohydrate source in ruminant feed since the in-vivo digestibility by the cow increased from 56.85 % to 76.54% (P < 0.001). The treated rice straw could also be used for commercial process such as the generation of Single Cell Protein. Growth of Sporotrichum pulverulentum on treated rice straw gave a protein product of 24.41 % as compared to 3.8% on untreated rice straw.     

A new approach to intermittent charging of valve-regulated lead-acid batteries in standby applications

For many years, intensive research has been undertaken to increase the life of valve-regulated lead-acid (VRLA) batteries. Overcharging results in excessive temperature in the battery, which degrades the chemical composition of the electrolyte. When the battery reaches the end-of-charge state, the energy being supplied to the battery is no longer consumed in the charge reaction and this additional energy is dissipated as heat within the battery. At this point, the oxygen cycle accelerates, which leads to temperature rise inside the battery. State-of-the-art control technology is required to control the charging of the battery and prevent the battery going into thermal runaway. This paper discusses the charging strategies for VRLA batteries in standby applications. Intermittent charging decreases the continuous overcharge which arises in the case of float charging. The charging regime used in intermittent charging must ensure the full recharge of the battery. This paper describes a new efficient method of charging batteries employing an intermittent charging technique called "Interrupted Charge Control." Laboratory tests and results are presented.     

Artificial neural network-driven federated learning for heart stroke prediction in healthcare 4.0 underlying 5G

In recent years, smart healthcare, artificial intelligence (AI)-aided diagnostics, and automated surgical robots are just a few of the innovations that have emerged and gained popularity with the advent of Healthcare 4.0. Such technologies are powered by machine learning (ML) and deep learning (DL), which are preferable for disease diagnosis, identifying patterns, prescribing treatments, and forecasting diseases like stroke prediction, cancer prediction and so forth. Nevertheless, much data is needed for AI, ML, and DL-based systems to train effectively and provide the desired outcomes. Further, it raises concerns about data privacy, security, communication overhead, regulatory compliance and so forth. Federated learning (FL) is a technology that protects data security and privacy by limiting data sharing and utilizing model information of distributed systems to enhance performance. However, existing approaches are traditionally verified on pre-established datasets that fail to capture real-life applicability. Therefore, this study proposes an AI-enabled stroke prediction architecture consisting of FL based on the artificial neural network (ANN) model using data from actual stroke cases. This architecture can be implemented on healthcare-based wearable devices (WD) for real-time use as it is effective, precise, and computationally affordable. In order to continuously enhance the performance of the global model, the proposed FL-based architecture aggregates the optimizer weights of many clients using a fifth-generation (5G) communication channel. Then, the performance of the proposed FL-based architecture is studied based on multiple parameters such as accuracy, precision, recall, bit error rate, and spectral noise. It outperforms the traditional approaches regarding accuracy, which is 5% to 10% higher.