There are various types of digital content protection techniques available in the market. Effectively securing the content is the main challenge as we have various advance tricks & methods available that can easily hack our premium content. So, delivering the content like OTT & Digital TV services to end-user makes advance data security a very interesting topic. This paper focuses on a key concept to use software-based encryption for Broadcast data security. This paper covers complete information about live data processing through broadcast equipment. The main point is securing the content through softwarebased encryption that will reduce the additional hardware needed to encrypt & decrypt the live content. The simplified approach of securing the content via open-source tools & platform makes the entire work interesting & cost-effective.
It is already reported in the literature that the performance of a machine learning algorithm is greatly impacted by performing proper Hyper-Parameter optimization. One of the ways to perform Hyper-Parameter optimization is by manual search, but that is time-consuming. Some of the common approaches for performing Hyper-Parameter optimization are Grid search Random search and Bayesian optimization using Hyperopt. In this paper, we propose a brand new approach for hyperparameter improvement i.e. Randomized-Hyperopt and then tune the hyperparameters of the XGBoost i.e. the Extreme Gradient Boosting algorithm on ten datasets by applying Random search, Randomized-Hyperopt, Hyperopt and Grid Search. The performances of each of these four techniques were compared by taking both the prediction accuracy and the execution time into consideration. We conclude that the Randomized-Hyperopt method is the best performer when compared to that of the other three conventional methods for hyper-parameter optimization of XGBoost.
The objective of developing a recommender system is to aid users by recommending products that might be of interest to them. In this, the collaborative filtering technique is one of the widely used methods where similarities are calculated among users/items to provide personalized recommendation. In order to calculate the similarity, various similarity measures are used. Most of these similarity methods do not perform satisfactorily in the presence of cold start users. A user is considered cold start if he/she has rated less than twenty items. In case of such users, the minimum available ratings data has to be utilized to recommend items. To resolve this problem, we propose a new similarity measure based on both City Block (CB) and Jaccard measure (CBJ). The co-rated items are considered by City Block while Jaccard measure considers the common items for similarity computation. Thus, when both of these measures are combined, they consider all the co-rated and common items. The main advantage of using CBJ is the reduced computational complexity involved in finding the similarity as compared to other similarity methods. To validate CBJ, we conduct experiments on Film Trust and MiniFilm data sets. The recommendation results on Film trust data set having 872 cold start users out of1508 users and MiniFilm data set having all the 55 cold start users reveal that the proposed CBJ method outperforms other existing methods.
This paper aims to convert the BTS/Node B used in the GSM/WCDMA communication into a smart BTS/Node B using IoT and thereby aims to achieve cost effective method to provide uninterrupted wireless communication. The proposed method makes the BTS/Node-B compliant for IoT and it control the power fed to BTS / Node B through Internet. This smart system is also capable of obtaining several system alarms from BTS and the power plant to update the operation and maintenance control terminal. The experimental results prove that system is cost effective and a noteworthy development in the telecommunication sector.
India is rapidly developing country in the world. As a growing economy it is important to manage its waste. In India on an average 64 million tons of detritus is produced which ranks 5th in global scenario. Now a days in many localities garbage is thrown arbitrary and roads are seen with full of litter. These unplanned things causes many problems sometimes it may cause hazardous diseases. It necessitates a management system that will curb this issue and has a complete observation on the detritus. The intention towards this paper mainly focused on the monitoring and tracking of garbage present in our ambience. In addition to that the data is sent into IOT based cloud platform for real time monitoring. After reaching the end value of garbage in the dustbin the alerts are sent directly to the municipal corporation through GSM module. If any fire occurs in the dustbin, it will get alert through buzzer.
Management of containers and carriers in a supply chain that spreads across different intermodal legs of ocean, land, river, rail and air transport is a challenging task in the shipping industry. During the intermodal phase, the triangulation of containers or carriers is a process that is sought to minimize cost by saving a possible transport leg. In this paper, we discuss an optimal triangulation process of containers carried by trucks in an intermodal transport network. We are addressing a specific triangulation process for the trucks engaged in import drops or export pickups of containers such that they can be effectively reused for the next export pickups or import drops in locations within a neighbourhood. We propose a mathematical model to address this problem in the framework of minimum cost network flows. Further, we introduce a heuristic method using the successive shortest path algorithm for the proposed model. The model is analyzed using data from current shipping networks of one of the major shipping industries for its North America database.
The recommendation system uses prior obtained information about the user to present user inteseted data. Personalized results aim to provide relevant information to the user based on the user's basic information or activity with the system. The user's basic information can be modeled into a user profile using ontology. Ontology is the systematic representation of various entities in a domain and the relationships between them. In this paper, we aim to present the conceptual model for a job recommendation system that uses ontology-based user profiles. The system collects basic information and models into a user profile. The dynamic aspects such as favorite jobs list and recently viewed jobs are then used as a source of data for the system. The recommendation algorithm works on the input given to present the list of relevant jobs to the user.
Soilless agriculture, hydroponics can be implemented efficiently with a Controlled Environment Agriculture System (CEA). The technological progress and improvements in smart farming have provided a platform for successful deployment of CEA. With more and more advances, the use of complex mathematical models by the hardware software interfacing, artificial intelligence and adaptive data analysis are providing the CEA with versatile design and control strategy to implement the broader level of automation. The review is an attempt to highlight the different hydroponic techniques their pros and cons in building an economic system. This study reviews various physical and environmental variables that influence the plant growth for the sustainable and efficient farming system. This research also highlights the methodologies that are used to automate, monitor and control the parameters for optimal plant growth. The research ultimately proposes the prediction models using machine learning techniques to understand the correlation analysis with plant growth dynamics. Finally, the research also focuses on the challenges that are to be identified while integrating smart farming in a CEA to minimize the energy inputs, enhancing productivity for higher quality crops.
Mobile Cloud Computing (MCC) is a smart way to reduce IT cost, provide services anywhere, and any time. Encrypting the outsourced data is one way of enhancing the privacy from the owner point of view. Mobile devices consumes more resources when workload is heavy. In order to reduce, computation, communication and searching time of related files based on keyword query “An Efficient Search Scheme over Encrypted Data with Indexing on Mobile Cloud” (SSEIM) is proposed. The SSEIM uses the keyword-based search that includes Order Preserving Encryption (OPE) to order the encrypted data and makes use of indexing the record to search the related files. Thus, the proposed scheme is more efficient with respect to retrieval time as compared to the existing scheme TEES (Traffic and Energy saving Encrypted Search).
The Web is overloaded with information and its exploration and retrieval is quite tedious is a cumbersome task. In the era of Semantic Web, there is a requisite of Semantic strategies for recommending webpages. In this paper, a strategic semantic paradigm for Ontology Driven Semantic search has been proposed. The proposed scheme incorporates a set expansion mechanism for interdomain exploratory semantic search. The proposed scheme for searching semantically computes concept similarity between concepts from most similar domains, and the set expansion is based on these concepts. The ontologies are visualized using a triple store and personalization has been imbibed into the proposed scheme. The initial step requires determination of the domains and scope of the ontologies, followed by identification of classes, defining instances of each class and the relationship among them. Multiple user profiles will be created to personalize the search results according to the search history of each user. Through this model, the problem of irrelevant search results is reduced, and in the process, reduces the probability of going through numerous results as in case of a normal search engine. Also, the search engine is made scalable to any dataset irrespective of its content. An overall F-measure of 96.64% has been achieved.
IoT today is expanding at a rapid rate. The number of connected devices is expected to hit 70 billion by 2025 and will increase further at an exponential rate. Additionally, the surge in IoT surveillance devices makes manual human monitoring impossible and presents a pressing need for automated monitoring. Thus, we present a paper that aims to detect anomalies from surveillance videos and identify the point and duration of anomaly occurrence in the video as well. We propose a novel multi-input neural network incorporating spatio-temporal features and dense flow features to train our dataset. We show the ability of our model to capture anomalous content better than baseline approaches. The low false alarm rate shows the resistance of our model to normal videos, despite containing a lot of motion. The introduction of our new standardized dataset also opens scope for further research in this field.
Stations for observing the weather are built, to collect meaningful and quantifiable data from the weather conditions in an expanse. The ever-changing weather makes it a concern for us, to make certain, that the weather conditions in an environment today are monitored and is of much significance. In this paper, a conceivable solution for an IoT based climate forecast system. The design implemented, primarily makes use of a Raspberry Pi 3 Model B, some sensors, and a weather forecast algorithm. Air pressure and temperature recordings values are used to predict the weather. One of the major goals was to discover an economical and purposeful solution for online and instant weather observing and prediction system for an expanse, and were able to carry that out by using the sensors in combination with the Raspberry Pi.
Twitter is the social media platform for real-time broadcasting of information on world events. The microblogging site contains and continues to generate huge amounts of data along with the growing breadth of a geographically diverse user base. Qualitative analysis of this enormous data will require substantial effort on information filtering to successfully drill down to relevant topics and events. This paper presents an automated learning system for trends in twitter to generate a recommendation system for users to understand the contexts in a particular trend. We have devised a framework using Apriori Algorithm and Named Entity Recognition on learned twitter trends. The paper also presents schemes for visual representation of the results using concept hierarchies.
Battery swapping has attracted attention from transportation companies, Original Equipment Manufacturers (OEM) and government authorities as a thriving solution to enable faster adoption of electric vehicles in the recent days. However, in order to facilitate a large scale electric vehicle operation that allows multiple independent stakeholders to engage in various different aspects of battery swapping operations, multiple challenges are required to be solved in order to ensure sustainable business models for all stakeholders. In this paper, we explore how the operations of a battery swapping station could be optimized to maximize it’s profit, which depends on a trade-off among the revenue earned from swapping operations and the cost incurred due to charging. We propose reinforcement learning based charging model that takes into account, this trade-off and adapts to the incoming vehicle arrival rate. We present simulation results to show that the proposed mechanism achieves higher profits than a greedy mechanism.
We propose a new fuzzy clustering algorithm by incorporating constrained class uncertainty-based entropy for brain MR image segmentation. Due to deficiencies of MRI machines, the brain MR images are affected by noise and intensity inhomogeneity (IIH), resulting unsharp tissue boundaries with low resolution. As a result, standard fuzzy clustering algorithms fail to classify pixels properly, especially using only pixel intensity values. We mitigate this difficulty by introducing entropy that measures constrained class uncertainty for each pixel. The value of this entropy is more for the pixels in the unsharp tissue boundaries. Apart from using the fuzzy membership function, we also define the similarity as the complement of a measure characterized by a Gaussian density function in non-Euclidean space to reduce the affect of noise and IIH. By introducing a regularization parameter, the trade-off between the fuzzy membership function and class uncertainty-based measure is resolved. The proposed algorithm is assessed both in qualitatively and quantitatively on several brain MR images of a benchmark database and two clinical data. The simulation results show that the proposed algorithm outperforms some of the fuzzy-based state-of-the-art methods devised in recent past when evaluated in terms of cluster validity functions, segmentation accuracy and Dice coefficient.
The Lensless Smart Sensor (LSS) developed by Rambus Inc. uses a revolutionary new approach to optical sensing by capturing information rich data in a tiny form factor. This paper experimentally verifies the superior quality of the new and improved version of the LSS as opposed to its earlier version. Also the paper examines the performance of the new sensor in 3D single point tracking. Sensor characteristics and accuracy are assessed which prove that the new sensor has considerable improvement and is able to track a point source down to millimeter level accuracy.
The ever-degrading quality of crops and excessive cost of organic fruits and vegetables in the market has put us all in a conundrum. A solution for this is setting up a kitchen garden. This paper presents us with a solution to help maintaining a kitchen garden having features like location and time-based crop prediction, live sensor data monitoring and irrigation using the sensor data. This paper also presents a mobile application which has an E-commerce platform for buying and selling the obtained yield. This paper emphasizes ways and means through which like-minded people can interact using the community ecosystem.
Modern computer threats are more complicated compared to the past. The noteworthy problem faced by many network enterprises mostly is from bots. A network of private computers infected with malicious software and controlled as a group without the owner's knowledge is known as botnet. We propose a new algorithm which uses backtracking approach to detect the bots in a network. This paper mainly focuses on bot properties and its behavior in the network. The performance metrics which we considered are response time, delay, network traffic, packets dropped and flood packets.
