Search

Software testing is the method of executing the program with the intension of finding the errors. This process is very complex and time intensive and costly. Automation of software testing process has turn out to be very workable solution in software engineering. For this purpose, machine learning algorithm is being used. We chose 48 primary research studies, tried to explain each machine learning-based approach in software testing. It gives an overview of how the input that is available data given to the model for learning and then how it is helping in giving predictions. We discovered that machine learning methods were mostly employed for test case development, refining, and evaluation. Machine learning has also been used to evaluate the production of test oracles and predict the cost of testing procedures.

Customer churn prediction in telecommunication industry is a very essential factor to be achieved and it makes direct impact to customer retention and its revenues. Developing a good and effective churn prediction model is very important however it is a time-consuming process. This study presents a very good review of customer churn, its effects, identification of its causes, business needs, methods, and all the techniques used for churn prediction. On the other hand, this study provides the best understanding of the telecom dataset, datasets used by past researches and features used in different researches. Also, this study shows the best techniques used for the churn prediction and describes all performance measures used in the churn prediction models. In this study a wide range of researches are added from the year 2005 to 2020. It includes variety of methods proposed by past researches and technologies used in these researches. At the end, a state of art comparison is added that gives a very good and meaningful comparison of past researches. The study indicates that machine learning techniques are mostly used and feature extraction is a very important task for developing an effective churn prediction model. Deep learning algorithm CNN itself has the capability of feature extraction and establish itself as a powerful technique for churn model, in particular for large datasets. For performance ‘Accuracy’ is a good measure however measuring performance only with ‘Accuracy’ is not sufficient because on small datasets accuracy is more predictable and will be the same. With Accuracy, researchers also need to look at other performance measures such as confusion matrix, ROC, precision. F-measure etc. This study assures that new researchers can find everything regarding their churn prediction model requirements at one place. This study provides a comprehensive view by extensively detailing work which has happened in this area and will act as a rich repositorory of all knowledge regarding churn prediction in telecom sector.

Today in every industry weather, it is ISP, IT products, social network or mobile services there is the problem of customer churn (Customers changing their services from one service provider to another). However, in telecommunication the customers churning very frequently. As the market in telecom is fiercely competitive, in that case, companies proactively have to determine the customers churn by analyzing their behavior and try to put effort and money in retaining the customers. In this proposed model, two machine-learning techniques were used for predicting customer churn Logistic regression and Logit Boost. Experiment was carried out in the WEKA Machine-learning tool, along with a real database from an American company Orange. The result were shown in different evaluation measures.

Digital images are being used as a rich source of information in the present digital era but the advancement of digital cameras & mobile phones and a rapid growth of image tempering software nowada...

Vehicular Ad-hoc Network is one of the challenging networks which combine different features of the different network in one communication form. This kind of network provides communication without setting up any infrastructure, configuration or the network topology. This research work provides an effective model to optimize communication in the Hybrid Vehicular Network. The model has achieved effective inter-region and intra-region communication with consideration of handoff issues. This work is focused on the optimization of the handoff process in a critical vehicle ad-hoc network. The rule formulation on various aspects is applied to identify the vehicle type, controller strength and the movement aspects. The decision is taken by the collective parameters for optimizing the Handoff. The model is applied to the highway scenario with the placement of fixed controller devices and the simulation is done on the NS2 environment with a different configuration. The result of simulation represents a highway scenario with RSU placement as the controller with integration to the communication technology. The communication can be performed between two nodes present in the same coverage area.

In this paper, a clustering algorithm named ICRA, which is based on Breadth first search approach has proposed. In this algorithm a new robust criterion NCN has introduced for deciding which vertex processing first from the list of vertices which are not present in any cluster. It efficiently mines the ordered sequences and update as well. This work is useful in community mining in social network analysis. The proposed algorithm is inspired by CRA algorithm, which does clustering twice. The proposed approach suits signed social networks too, and effectively mine negative vertices. In addition, this algorithm improves the predictive performance; especially for negative linked inter-communities datasets hence increases the accuracy when tested with the Gahuku - Gama dataset.

In the present manuscript, an adaptive evolutionary multiple sequence alignment algorithm is proposed that uses a combination of consensus and SP-score methods. The algorithm searches intermediate pairwise consensus strings that are used to identify the final consensus string for a given set of DNA/RNA/protein sequences. The proposed algorithm is an extension of MPSAGA algorithm that uses positional matrix representation of sequences. An empirical study was performed in the present work to compare the proposed algorithm with the other three contemporary ClustalW, TCOFFEE, and MUSCLE algorithms on the four datasets. The overall observations from the various experiments revealed that the proposed algorithm outperforms than the other algorithms tested in aligning multiple sequences with an average increase of 0.03% in alignment length by inserting 0.02% increased number of gaps.

Due to large scale use and many applications of cloud maximum data, owners upload the information at cloud space to save time and local disk space. Here, the authors provided a TBMKRS (Tree-Based Multi-Keyword Rank Search) scheme which also supports dynamic update (insert/delete) and verifiability of encrypted information upon cloud. For the evaluation of performance and analysis of result, the Enron data set have been used by the authors.

The brain tumor is a disease that affects or harms the brain with unwanted tissues. This is very difficult to detect brain tumor tissue from whole brain. Early detection of tumor is very important to save patient's life. Detection or segmentation techniques are used to detect and segment the brain-tumor region from the MRI images of brain and it is very useful method in recent days. In medical, magnetic-resonance-imaging is a tough field in image processing because accuracy percentage must be very high so doctors could get proper idea about diseases to save patient's life. Some MRI images have been taken as inputs data. The brain-tumor segmentation process is performed for separating brain-tumor tissues from brain MRI images, The MRI images should be filtering such as with the median filtering technique and skull stripping should be done in pre-processing, the thresholding process is being done on the given MRI images with using the watershed segmentation method. Then at last the segmented tumor region is obtained. And then in other phase features extracted by GLCM methods using MATLAB software. Then, the some images have been classified using support vector machine (SVM), this system obtained with the average accuracy of 93.05%. Which is quite better than other conventional models.

In the beginning of the information age Web 1.0, whatever content which was available on the Internet was used only for the purpose of reading. But, things changed after Web 2.0 as people started sharing what they think about a particular product or an object. This related information generated by user became of paramount importance when it started affecting the way other people think of the same objects and this formulated the need of opinion mining or sentiment analysis by different organizations. Opinion mining or sentiment analysis is the process of determining the positive or negative inclination of a piece of written information. Feature-based opinion mining works upon the individualistic features of an opinion. This paper proposed a robust algorithm for handling the issues of double intensifiers when expressed in an opinion especially in negative opinion by performing feature-based opinion mining on movie reviews and discusses over future aspects of the same.

In these days enterprises are using cloud space to keep their data. This practice saves the cost of hardware purchasing and maintenance as well as infrastructure cost. Although privacy and security of the data are the important issues because the cloud server is either un-trusted or semi trusted. Various methods and schemes have been used to achieve the objective of data security and privacy. To keep data safe and secure, data is encrypted before uploading at cloud space. Therefore to search particular information at cloud space the plaintext keyword searching method is obsolete. The new method for searching the data upon cloud is searching upon encrypted data. The authors proposed here a multi keyword rank search scheme, which returns top most relevant documents as an output of search. The tf x idf model as well as vector space model is used to form the table that keeps the index and queries. For encryption of index and query kNN algorithm can be used. The authors have formed a tree known as M way balanced search (MBS) for search and indexing in a better way. The authors also proposed a dfsm (depth first searching method) scheme for multi key synonym rank search. The certainty and efficiency of dfsm is scalable but MBS tree is taking sub linear time complexity. For calculating the complexity computational methods were used.

Secondary metabolites are natural products produced by plants, fungi, and bacteria. After their discovery and due to the revolution in the pharmaceutical industry, secondary metabolites such as penicillin, streptomycin, and cyclosporine were manufactured in bulk for human welfare. The secondary metabolites are synthesized by the specialized enzymes encoded by coregulated genes located on the biosynthetic gene clusters (BGCs). There are two methods to identify fungal secondary metabolites and their corresponding BGCs: manual methods or in silico methods. The manual methods are time-consuming and complex; conversely, in silico tools are relatively simple, and promising. Most of the in silico tools identify previously known biosynthetic gene clusters by matching the gene sequences to the already available secondary metabolites gene data sets. Some of the in silico tools such as SMURF, ClusterFinder, and Grape can identify known and unknown biosynthetic gene clusters in the submitted genome. Due to the availability of several online tools, it is crucial to know the key functions of each tool to solve a research problem. Therefore, in this chapter, a comprehensive survey of currently available in silico tools has been made that can identify biosynthetic gene clusters in various fungal genomes. The key functions of in silico tools, namely SMURF, antiSMASH, PRISM, SMIPS, SeMPI, DEREPLICATOR +, CASSIS, MIDDAS-M, NP.searcher, ClustScan, CLUSEAN, FunGeneClusterS, ClusterFinder, EvoMining, MIPS-CG, Pep2Path, GRAPE, GARLIC, NRPSpredictor, and SMBP, are discussed, along with the official uniform resource locator (URL) for each tool. Schematic workflow for the identification of the biosynthetic gene cluster has also been proposed, which is further extended as a decision chart in this review.

In recent trend biometric authentication is been used extensively to authenticate the person remotely. So, it is necessary to secure the sensitive secret data by encryption, before transmitting over medium. This paper proposed a stout, highly secure authentication technique based on semantic segmentation, Triple key chaotic encryption and data hiding. If a user wants to be authenticated remotely, then firstly image is segmented to extract head and body parts through semantic segmentation. Second, take A’s fingerprint and encrypt it using Triple key chaotic encryption technique. Furthermore, qualified significant wavelet tree is used to place the encrypted signal in the most significant wavelet coefficient of image. Finally, Inverse Discrete Wavelet Transform is applied to obtain stego image. The output image is further compressed before transmission over wireless medium using Set Partition in Hierarchical Tree compression technique. This technique has good resistance against lossless transmission and compression. The performance measures are Peak signal to noise ratio, correlation index, information entropy and compression ratio.

Data security is one of the dominant essential problems these days where data is delivered from one to other location with fast values. Image encryption is the technique of encoding image in like a way that detectives or hackers cannot understand it, but that permitted groups can. In this work, we have proposed an efficient approach of image encryption with the help of combination of Arnold cat map, zigzag transformation and pixel scrambling technology in fractional Fourier transform and compared with the previous existing techniques. Some performance parameters such as peak signal-to-noise ratio (PSNR), mean squared error (MSE) and normalized cross-correlation are also measured. All the simulations are carried out in MATLAB simulation tool.

The Requirement of system capacity is increasing due to the rise in the number of computer users. This can be achieved by optical communication using WDM technology. Using WDM technology multiply many carrier signals onto a single optical fiber using different wavelength to boost up capacity. Dispersion effects the transmission of the signal along an optical fiber. In the proposed model, three techniques for dispersion compensation (per, post, and mix) has been designed using dispersion compensation fiber (DCF) in 8 channel WDM. According to this topology with the help of 5 Gbps transmission data rate model is compared non-return-to-zero (NRZ), return-to-zero (RZ) pulse in which signal mode fiber (SMF) has been considered using the length of 120 km. The results such as Q-factor and bit error rate (BER) represent pre-compensation scheme using RZ gives better results and also output power at EDFA calculate using optisystem 7.0 software.

The proposed algorithm is a novel matrix-based global pair-wise sequence alignment with a de novo sequence representation. Needleman–Wunsch, noblest, Emboss-Needle, ALIGN, LALIGN, FOGSAA, DIALIGN, ACANA, MUMmer, etc. are few other algorithms that are most commonly used for global pair-wise sequence alignment. Needleman–Wunsch algorithm is one of the most popular algorithms that provides the best possible pair-wise sequence alignment, but the algorithm output is associated with high time and space complexities. To resolve these complex issues, researchers have proposed several algorithms to reduce time and space complexities in the pair-wise sequence alignment. Most of these algorithms provide solutions, but compromise the optimal result in favor of plummeting time and space complexities. An attempt has been made in the present research to develop MPSAGA and a completely unique positional matrix (PM) based sequence representation to deal with the time and space complexities without compromising sequence alignment results (MPSAGA is in public domain available at https://github.com/JyotiLakhani1/MPSAGA ). A benchmarking of the proposed algorithm has also been performed with other popular pair-wise sequence alignment algorithms with and without positional matrix-based sequence representation. The use of an integer instead of string data type and exclusive clustering method in MPSAGA with positional matrix based sequence representation resulted in a noteworthy reduction in the memory usage (space) and execution time in the pair-wise alignment of biological sequences.

Active contour models, colloquially known as snakes, are quite popular for several applications such as object boundary detection, image segmentation, object tracking, and classification via energy minimization. While energy minimization may be accomplished using traditional optimization methods, approaches based on nature-inspired evolutionary algorithms have been developed in recent years. One such evolutionary algorithm that has been used extensively in active contours is the particle swarm optimization (PSO). However, conventional PSO converges slowly and gets trapped in local minimum easily which results in inaccurate detection of concavities in the object boundary. This is taken care of by using proposed multiswarm PSO in which a swarm is set for every control point in the snake and then all the swarms search for their best points simultaneously through information sharing among them. The performance of the multiswarm PSO-based search process is further enhanced by using dynamic adaptation of the inertia factor. In this paper, we propose using a set of fuzzy rules to adjust the inertia weight on the basis of the current normalized snake energy and the current value of inertia. Experimental results demonstrate the effectiveness of the proposed method compared to conventional approaches.

In the today’s age of the information technology the data is required to share in every respect whether structured or unstructured. In this paper we have proposed the search associate based memory approach which performs the comparison in the faster way and also to securing shared the file using the photo based grid and SHA verification of the file.

Wireless Sensor Networks (WSN) for communication between the nodes uses a different type of protocols. This paper proposed improved Power-Efficient Gathering in Sensor Information System (PEGASIS) chain based routing protocol. The protocol are sensor nodes connected with next near nodes and selects the leader node for communication to the base station but this causes some problem when used for a long distance between nodes and base station. Such problem causes large transmission delay, selection of a leader, scalability, and unbalanced energy consumption. For enhancing the performance improved PEGASIS protocol is proposed which provides better energy efficiency and good lifetime sensor networks as compared to the unique PEGASIS protocol.

This paper presents the leaf shape extraction for the classification of plant. Leaf shape is an important parameter which are frequently used to identify and classify the plant species. Besides many challenges occurred in plant classification, it is very useful technique to study the livestock systems, conservation and ecology. The camera captures the image followed by segmenting the required leaves from the image and then a single leaf is segmented using OTSU's segmentation. A new approach is used to extract 14 features from the leaf using shape detector and classifying 16 different kind of plants with SVM classifier. Accuracies ranging from 89% to 91% are acquired which are similar to the earlier reported work.

A variety of image encryption algorithms have been initiated to prevent personal and confidential images from various security attacks. This paper is proposing an encryption and decryption idea on colour image by using component based Partial Random Phase Encoding (PRPE) and Fractional Fourier Transformation. The encoding converts the input RGB colour image into HSV (Hue, Saturation, Value) format. In the formation of HSV, the order of Hue and Saturation contains phase masking of higher same order while V consists of lower order than H and S. A Fractional Fourier transform pretend as a key which offers extra degree of security in different orders. By using correct keys and Fractional orders, primary images can be recovered. The HSV image has reverted back to the RGB format again in decryption method. The level of encryption of various techniques compute the robustness of image and it is verified by mean square error (MSE) as well as peak signal to noise ratio (PSNR) value. Simulation results of versatile analysis shows that this algorithm provides high security and resourceful for colour image.

Wireless sensor networking is an emergent technology that attracted a wide range of potential application in both civilian, military areas and modern application for monitoring environmental and target tracking. The major problem of wireless sensor networks is very small lifetime of sensor networks and it needs very high-level maintenance because the sensor nodes are equipped with low power battery. Probability model that the forms cluster head is unbalanced and unreachable. It is always desired to maximize the overall performance of the nodes. In this paper, proposed Sensitive and reachable cluster head (SRCH) routing protocol. In this protocol cluster head is selected based on ratio between the current energy level and an initial energy level, and multiplies by the root of number of neighbor nodes. The simulation results of SRCH protocol, is compared with LEACH and TEEN protocol and found more efficient with higher lifetime.

Traffic sign recognition (TSR) is a complex system to recognize many traffic sign, but sometimes it does not gives the correct result. To solve the problem of this system introduced self-organization Feature map (SOFM) neural network to classify traffic sign and feature extracted by Discrete Cosine Transform (DCT) of sign images. The proposed scheme is tested under manual database and shows the effectiveness by recognition rate is high and extraction time is less as compared to literature works.

The complete or partial obstruction of light by an opaque object forms a shadow. It is an undesirable effect, as it not only degrades the image quality but also limits the information conveyed by that image. Thus for correct interpretation of the image, shadow detection and removal becomes a necessary step in image processing. Many works have been carried out in this field which can successfully detect and remove the shadow from an image, for example gamma correction, histogram matching, correlation techniques etc but it is still a challenging task to do the same automatically. This paper presents a full automatic gamma correction technique to eliminate the shadow from an image.

Video Watermarking is an important feature of the data security and the secure data transmission of the secrete data. Video watermarking can be very useful when the data size is large and also if the data is in complex format. In this paper the algorithm is based on the frame data rate and single frame data hiding is concentrated. The algorithm developed can be altered according to the frame rate and also the data hiding can be done over the selected data frames, the algorithm is firstly triggered only for the 1280×960 frame size though it can be of any size but because of the available hardware and resources this size has been selected.

Image watermarking is an important and a complex task if done with the high security parameters, in this paper the main focus is over the raising the complexity level of the hiding algorithm, hence there have been as many as three proper major equations which are used to hide the secrete text data inside an image. along with three major and minor equations there have been few more complex steps like dwt which have been used to keep the watermarked image quality high on different parameters. image watermarking is a significant and compound feature of information security. digital watermarking is the procedure of embedding information into a digital signal in a manner that is not easy to remove. watermarking is an hidden text embedded within an image to explain authenticity or verification of holder ship. a digital watermark is a blueprint of bits embed into a digital image, audio or video file. in this paper the center of attention is on the text data watermarking to achieve the preferred outcome there are several techniques which are going to be discussed in this paper. the techniques which are going to be discuss.

This paper deciphers the Facial Expression Recognition (FER) using Gabor Filter and Artificial Neural Network (ANN), extracts the facial expression using Gabor filter and then classify the facial expressions using the multi-layer artificial neural network. Recognizing facial expressions of human beings in an image processing by a computer is an interesting and challenging research work. This paper is based on detection and classification of facial emotion expressions. At first, we design an algorithm to detect the face reason image in the whole image using Viola-Jones detection algorithm, then by using Gabor filter extracts the facial features in the spatial domain. The Gabor filter is used to capture the whole frequency spectrum in all directions. And then extract meaningful facial features using Gabor Filter. Finally, they have been successfully classified the facial expressions using the extracted Gabor features of face image used as an input to the Artificial Neural Network classifier. The experimental results on database images of JAFFE show the robustness and better recognition rates of the proposed approach.

The undesirable local change in the illumination intensity observed in a captured image often due to some opaque object hindering the light path is termed as shadow. Apart from resisting the image to deliver the high quality content, it also reduces the efficiency of various next level image processing applications such as object detection, feature extraction etc. This paper proposes a method to recoup a shadow recovered image. The method involves the detection of shadow which also includes morphological filtering to rule out the ambiguities and then eliminating the detected shadow. This paper adopts Gamma correction technique for shadow removal. The experimental results verify the efficiency of this method by comparing it with two other methods. The parameter Structural Similarity Index Metric is also used to gauge the accuracy of the three methods.

An edge is a set of connected pixels lying on the boundary between two regions in an image that differs in pixel intensity. Accordingly, several gradient-based edge detectors have been developed that are based on measuring local changes in gray value; a pixel is declared to be an edge pixel if the change is significant. However, the minimum value of intensity change that may be considered to be significant remains a question. Therefore, it makes sense to calculate the edge-strength at every pixel on the basis of the intensity gradient at that pixel point. This edge-strength gives a measure of the potentiality of a pixel to be an edge pixel. In this paper, we propose to use a set of fuzzy rules to estimate the edge-strength. This is followed by selecting a threshold; only pixels having edge-strength above the threshold are considered to be edge pixels. This threshold is selected such that the overall probability of error in identifying edge pixels, that is, the sum of the probability of misdetection and the probability of false alarm, is minimum. This minimization is achieved via particle swarm optimization (PSO). Experimental results demonstrate the effectiveness of our proposed edge detection method over some other standard gradient-based methods.

In this document, a protocol LEACH-1R meant for energy efficient & mobility aware congestion control is provided for MWSN. A LEACH-1R that transforms the LEACH by making use of first clustering round & performing the inter-cluster choosing of new CHs (cluster head) rather than aggregated reclustering. A MWSN is termed as a set of wireless mobile nodes & base station that generates an ad-hoc network. Such network can be implemented over several kinds of fields, like underground & underwater. Each node is provided with a particular amount of energy & randomized mobilization. As the absorption of energy is considered to be a cardinal aspect, the protocols of Hierarchal type are able to cope up with this issue, as an illustration, LEACH protocol. LEACH-1R is considered to be an extension to LEACH protocol that improvise the lifetime of a network but the protocol also makes use of location of sensors for organization of network. Hence, it is not considered to be much effective in that field where node localization with the mobility is not possible. In this document, we tend to improvise the traditional algorithm for generating a latest protocol that is termed as LEACH-1R Decentralized for enhancing the lifetime of a network with applying the maps. Further, this protocol is put in contrast to the SMAC & TSEEC for the lifetime of network & dissipation of energy by making use of network simulator. In this experimentation, we tend to enhance the performance of packet delivery ration for a 50 MSN having 2 cluster & 100 MSN comprising 3 cluster.

Mesh-based multicast routing protocols in Mobile Ad Hoc Networks (MANETS) allow data to be sent to all the members of a network through multiple paths. As a result, these are more efficient in comparison to tree-based protocols, provided the links remain stable over a period of time. Otherwise this redundancy results in a waste of energy on links that eventually die out before complete data transfer. In this paper, a new protocol SLIMMER (Stable Links in Multicast Mesh and Energy Efficient Routing) has been introduced, which overcomes this problem by considering energy efficiency of the nodes as a decision criterion to improve link stability. Its performance has been compared with the popular mesh-based multicast routing protocol PUMA. Results show that SLIMMER outperforms PUMA in terms of packet delivery ratio (PDR) and energy consumption, under conditions of varying number of nodes, mobility, and simulation time.

In this printed material we intended to lessen the high operational state temperature of electro-thermally impelled micro-gripper. As a rule cases the operational state temperature of electro-thermally activated small scale grippers is high for the most part in the scope of thousand Kelvin. It is the greatest downside of electro-thermally incited smaller scale grippers for some application reason and it is extremely hard to diminish it as low as could be expected under the circumstances. In this paper we diminish the operational state temperature of electro-thermally incited micro-gripper from thousand Kelvin to 377 [K] with the displacement of 17.9 μm and for incitation parallel biasing component is utilized with 0–2 volt go.