Cries of infants can be seen as an indicator of pain. It has been proven that crying caused by pain, hunger, fear, stress, etc., show different cry patterns. The work presented here introduces a comparative study between the performance of two different classification techniques implemented in an automatic classification system for identifying two types of infants' cries, pain, and non-pain. The techniques are namely, Continuous Hidden Markov Models (CHMM) and Artificial Neural Networks (ANN). Two different sets of acoustic features were extracted from the cry samples, those are MFCC and LPCC, the feature vectors generated by each were eventually fed into the classification module for the purpose of training and testing. The results of this work showed that the system based on CDHMM have better performance than that based on ANN. CDHMM gives the best identification rate at 96.1%, which is much higher than 79% of ANN whereby in general the system based on MFCC features performed better than the one that utilizes LPCC features.
Cries of infants can be seen as an indicator of pain. It has been proven that crying caused by pain, hunger, fear, stress, etc., show different cry patterns. The work presented here introduces a comparative study between the performance of two different classification techniques implemented in an automatic classification system for identifying two types of infants' cries, pain, and non-pain. The techniques are namely, Continuous Hidden Markov Models (CHMM) and Artificial Neural Networks (ANN). Two different sets of acoustic features were extracted from the cry samples, those are MFCC and LPCC, the feature vectors generated by each were eventually fed into the classification module for the purpose of training and testing. The results of this work showed that the system based on CDHMM have better performance than that based on ANN. CDHMM gives the best identification rate at 96.1%, which is much higher than 79% of ANN whereby in general the system based on MFCC features performed better than the one that utilizes LPCC features.
CNNs are playing a vital role in the field of automatic speech recognition. Most CNNs employ a softmax activation layer to minimize cross-entropy loss. This layer generates the posterior probability in object classification tasks. SVMs are also offering promising results in the field of ASR. In this article, two different approaches: CNNs and SVMs, are combined together to propose a new hybrid architecture. This model replaces the softmax layer, i.e. the last layer of CNN by SVMs to effectively deal with high dimensional features. This model should be interpreted as a special form of structured SVM and named the Convolutional Neural SVM. (CNSVM). CNSVMs incorporate the characteristics of both models which CNNs learn features from the speech signal and SVMs classify these features into corresponding text. The parameters of CNNs and SVMs are trained jointly using a sequence level max-margin and sMBR criterion. The performance achieved by CNSVM on Hindi and Punjabi speech corpus for word error rate is 13.43% and 15.86%, respectively, which is a significant improvement on CNNs.
Data loss occurs due to crashing, correlated failure, logical failure, power outages and security threats. Several techniques (e.g. NoBackup, WARBackup and LocalRecovery) are being used to recover data locally. And, strongly consistent Cloud services (SCCS) must provide good performance and high availability. However, conventional strong consistency replication methods have the limitation of availability of replicated services when recovering huge amount of data across wide area links. There is a need for remote recovery mechanisms for high availability of service/data, because distributed nature of cloud infrastructures. To address these issues, the article proposes a hierarchical system architecture for replication across a data center, and employs the backward atomic backup recovery technique (BABRT) for local recovery and remote recovery for high availability of the cloud services/data. A mathematical model for BABRT is described. Simulation results show that BABRT reduces the storage consumption, recovery time, window of vulnerability and failure rates, compared to other recovery models.
The Smart Home is an environment that enables the resident to interact with home appliances which provide resident intended services. In recent years, predicting resident intention based on the contextual modalities like activity, speech, emotion, object affordances, and physiological parameters have increased importance in the field of pervasive computing. Contextual modality is the feature through which resident interacts with the home appliances like TVs, lights, doors, fans, etc. These modalities assist the appliances in predicting the resident intentions making them recommend resident intended services like opening and closing doors, turning on and off televisions, lights, and fans. Resident-appliance interaction can be achieved by embedding artificial intelligence-based machine learning algorithms into the appliances. Recent research works on the contextual modalities and associated machine learning algorithms which are required to build resident intention prediction system have been surveyed in this article. A classification taxonomy of contextual modalities is also discussed.
Workflow technology is the core technology to realize business process modeling, process operation, monitoring and management, and ultimately realize business process automation. Workflow-based office automation system can separate code writing and the operation mode. When business processes need to be changed, there is no need to modify the program. Users only need to customize the workflow through the visual process customization mode. The workflow engine is the core of the whole workflow management system and the control center of the whole system. This article designs a workflow engine based on a relational structure, including the design of workflow engine class, functional component, interface and database. Finally, a flexible office automation system with customizable business processes is implemented.
Cloud computing is the most prevailing paradigm, which provides computing resources and services over the Internet. Due to immense development in services provided by cloud computing, the trend to share large-scale and confidential data on cloud has been increased. Though cloud computing provides many benefits, ensuring security of the data stored in cloud is the biggest challenge. The security concern about the data becomes main barrier for adoption of cloud. One of the important security aspects is fine grained access control mechanism. The most widely used and efficient access control scheme for cloud computing is Attribute Based Encryption (ABE). The Attribute Based Encryption (ABE) scheme provides a new technique for embedding access policies cryptographically into encryption process. The article presents an overview of various existing attribute-based encryption schemes and traditional access control models. Also, the comparison of existing ABE schemes for cloud computing, on basis of various criteria is presented in the article.
This proposed literature survey provides basic data regarding the first step of risk identification and analysis to achieve a secured infrastructure. The demand and risk are two parts of the same coin. The demand is directly proportional to the risk, but preventive control is inversely proportional to risk. The necessity of preventive control in any organization has increased because of the changes in logic, structure, and the type of technology applied to services that generate risks. Finally, the business increases along with technology, which creates risks and spreads over its infrastructure. We have to focus on protecting, detecting, correcting, verifying and validating the Unix file system. This survey article proposes and resolves the Unix file system by applying a hardening, re-configuration and access control mechanism up to the highest level of preventive control.
Modern information systems are expected to assist users with diverse goals, via exploiting the topical dimension (‘what' the user is searching for) of information needs. However, the intent dimension (‘why' the user is searching) has preferred relatively lesser for the same intention. Traditionally, the intent is an ‘immediate reason, purpose, or goal' that motivates the user search, and captured in search contexts (Pre-search, In-search, Pro-Search), an ideal information system would be able to use. This article proposes a novel intent estimation strategy; based on the intuition that captured intent, and proactively extracts likely results. The captured ‘Pre-search' context adapts query term proximities within matched results beside document-term statistics and pseudo-relevance feedback with user-relevance feedback for In-search. The assessment asserts the superior performance of the proposed strategy over the equivalent on tradeoffs, e.g., novelty, diversity (coverage, topicality), retrieval (precision, recall, F-measure) and exploitation vs exploration.
The use of Radio Frequency Identification (RFID) technology has attracted a huge attention from the supply chain business community. This is due to the use of RFID technology wide range of applications in the fields of logistics and supply chain management. This paper presents a brief overview of a simple industrial RFID system and then describe the basic concept of tag collision problem. Despite many useful applications, the RFID tag collision creates a major problem for fast tag identification process. There are different algorithmic solutions available to overcome tag collision problem in industrial supply chains. Based on binary search algorithm (BSA) of dynamic and backtracking, two variations of binary anti-collision search algorithms for tag identification processes are described in this paper. Simulation-based experimental results on the performance of these algorithms are evaluated when handling multiple RFID tags simultaneously. The backtracking binary search algorithm has obvious advantages in terms of tag identification process compared to the other two algorithms.
In the face of the rapid growth of data, the authors note the birth of new computer systems leading the management of databases. To ensure reliable handling of data which is bulky and heterogeneous, new systems of basic management data are differentiated from traditional ones by their linear scalability to store and manage data in parallel. Those systems are called NoSQL DB. Robotics database management systems are increasing. These systems ensure good storage of data and with big data analytics, a new approach demands new structures and methods for collecting, recording, and analyzing enterprise data. This journal article deals with the NoSQL DB which are the secret of the continual progression data that new data management solutions have emerged. The implementation of NoSQL for Robotics wrestles all data they acquire into a usable form because with the ordinary type of Robotics, people are facing big limits to find the best decision in real-time. The authors' proposed approach was based on a combination of some experimental studies in the same area.
Nowadays, in the field of intelligent transportation, data mining technology has been applied more and more widely. Data mining technology can find valuable data from amongst massive traffic data and analyze traffic conditions according to actual traffic conditions. In order to improve the management and control level of ITMS, effective information can be queried from various query conditions, models suitable for various traffic situations can be found, analyzed and predicted, and accurate information can be provided to traffic managers for decision-making. This article mainly studies the data mining algorithm in intelligent transportation, in order to provide practical reference for the application and research of urban traffic big data technology.
Cloud storage (CS) is gaining much popularity nowadays because it offers low-cost and convenient network storage services. In this big data era, the explosive growth in digital data moves the users towards CS but this causes a lot of storage pressure on CS systems because a large volume of this data is redundant. Data deduplication is an effective data reduction technique. The dynamic nature of data makes security and ownership of data as a very important issue. Proof-of-ownership schemes are a robust way to check the ownership claimed by any owner. However, this method affects the deduplication process because encryption methods have varying characteristics. A convergent encryption (CE) scheme is widely used for secure data deduplication. The problem with the CE-based scheme is that the user can decrypt the cloud data while he has lost his ownership. This article addresses the problem of ownership revocation by proposing a secure deduplication scheme for encrypted data. The proposed scheme enhances the security against unauthorized encryption and poison attack on the predicted set of data.
Gamification as a tool or procedure to add entertaining and motivating elements to usually non-entertaining environments such as schools or workplaces is becoming more and more popular. E-learning platforms like Moodle provide tools and sets of functions to add elements of gamification. An important factor, especially for education, is technology: individual achievements and progress can be recorded, measured, tracked and visualized and, therefore, identified and honored through bonus points, awards or rankings. This is where gamification can add some challenge and excitement to learning.
The identification, segmentation, and detection of the infected area in brain tumor is a tedious and a time-consuming task. The different structures of the human body can be visualized by an image processing concept, an MRI. It is very difficult to visualize abnormal structures of the human brain using simple imaging techniques. An MRI technique contains many imaging modalities that scan and capture the internal structure of the human brain. This article concentrates on a noise removal technique, followed by improvement of medical images for a correct diagnosis using a balance contrast enhancement technique (BCET). Then, image segmentation is used. Finally, the Canny edge detection method is applied to detect the fine edges. The experiment results achieved nearly 98% accuracy in detecting the area of the tumor and normal brain regions in MRI images demonstrating the effectiveness of the proposed technique.
Breast cancer is one of most dangerous diseases and more common in women. The early detection of cancer is one of the most key factors for possible cure. There are numerous methods of diagnosis amongst which: clinical examination, sonar and mammography, which is the best and more effective in detecting breast cancer. Detection of breast tumors is difficult because of the weak illumination in the image and the overlap between regions. Segmentation is one the crucial steps in locating the tumors, which is an important method of diagnosis of the computer. In this study, segmentation techniques are proposed based on; classic morphology and fuzzy morphology, and a comparison between them. The proposed methods were tested using the database of mini -MIAS, which contains 322 images. After the comparison the statistical results, it shows, the detection of tumor boundary with fuzzy morphology give the higher accuracy than the results in classic morphology. The accuracy is 60.69%, 58.61% respectively due to the high flexibility of foggy logic in dealing with the low lighting in the medical images.
This paper proposes a new algorithm to discriminate Arabic poems by inserting the Arabic poems texts and coding Arabic letters, extracting letters features depending on letters shapes to construct multidimensional contingency table, and analyses the frequencies of letters in the inserted texts statistically. The proposed coding and discrimination (CODIS) algorithm could be applied for different Arabic texts in any media. A sample of five poems for six poets was examined to implement (CODIS) algorithm Chi-Square statistic is used to determine the relation between the features and discriminate poems.
Recommender systems have been used successfully in order to deal with information overload problems in a wide variety of domains ranging from e-commerce, e-tourism, to e-learning. They typically predict the ratings of unseen items by a user and recommend the top N items based on user's profile. Moreover, the profile can be enriched further by using additional information such as contextual data, domain knowledge, and tagging information among others for improving the quality of recommendations. Traditional approaches have not been effective in exploiting these additional data sources. Hence, new techniques need to be developed for extracting and integrating them into the recommendation process. In this article, the authors present a survey on state of the art recommendation approaches their algorithms, issues and also provides further research directions for developing smart and intelligent recommender systems.
This article describes how it is impractical for a person to remember everything that they do on a day-to-day basis. To address this issue, an android location based reminder system SmartNotify that function on users' activities and points of interest are developed. SmartNotify automatically updates preferences of the user based on location behavior on a daily basis using validation from the stay detection algorithm. In addition, SmartNotify provides suggestions for the best locations that people visit frequently in the nearby area by making use of the DBSCAN algorithm and the Apriori algorithm. The utilization of data mining techniques in the android application makes the reminder application more efficient than the traditional way of notifying the user about their events.
In recent years, some sport clubs have adopted web forums for online discussions about planning training sessions, races, club problems, sponsors and supporters, equipment and so on. Mostly, these forums are closed, because some discussions about critical information must be permitted only to registered club members. Indeed, various members are connected into a network representing a forum community that participates in forming the internal structure of a club and has taken an increasingly important role in solving the club affairs. This has influenced the classic way of doing business with meetings or calls and has allowed the community to participate in forming plans. This article deals with the deep analytics of data acquired from a web forum of a small cycling club located in Slovenia for a period of five years. The purpose of these analyses is to identify various members of the forum's community, track the dynamics of events as found in this forum, and search for hidden relationships.
