Browse Items (1473 total)
Sort by:
-
A framework for integrating nested queries in natural language interfaces to databases
To translate Natural Language (NL) statements into Structured Query Language
(SQL) queries, different methods and systems were proposed in the past. This work presents a framework for automating the translation of Data Requirement
Specifications (DRS) given by enterprise Business Users in NL into SQL queries,
focusing on requirements that result in the generation of nested SQLs. The framework takes the business user’s DRS given in English as input and generates an initial query sketch by employing semantic parsing. This initial sketch is further refined and completed into a well-formed SQL by consulting the Database Schema. It performs the translation by combining NL processing techniques with Query Sketch generation methods and refines it by employing Repair techniques or extends it further. The framework suggests using Lambda expressions for intermediate representation and employs standard operations of Lambda Calculus for performing the required transformations needed for translation. Lambda Context Calculus (LCC) provides the operational semantics and the relevant methods needed in transforming NL statements into SQL, preserving the integrity and compositionality24 of expressions in every step of the translation. Though Lambda Calculus is found to be effective in representing the intermediate expressions and assists in performing the transformations that are needed for translating specific predicates into SQL, its inflexibility in combining parallel computations is a constraint. To represent clauses that are in parallel or are in
pipeline, and to perform the required transformations on the intermediate expressions involving these, more advanced programming constructs are needed. It also adopts functional programming techniques to deal with complex scenarios involving nested queries. This work recommends the use of some advanced language constructs, the Fixed-point Combinators11 and Monad Comprehensions20 for performing the required transformation at the intermediate language level and adopts functional programming techniques for the required syntactic sugar. -
A Framework for Integrating Nested Queries in Natural Language Interfaces to Databases
To translate Natural Language (NL) statements into Structured Query Language (SQL) queries, different methods and systems were proposed in the past. This work presents a framework for automating the translation of Data Requirement Specifications (DRS) given by enterprise Business Users in NL into SQL queries, focusing on requirements that result in the generation of nested SQLs. The framework takes the business user s DRS given in English as input and generates an initial query sketch by employing semantic parsing. This initial sketch is further refined and newlinecompleted into a well-formed SQL by consulting the Database Schema. It performs newlinethe translation by combining NL processing techniques with Query Sketch generation newlinemethods and refines it by employing Repair techniques or extends it further. The newlineframework suggests using Lambda expressions for intermediate representation and newlineemploys standard operations of Lambda Calculus for performing the required newlinetransformations needed for translation. Lambda Context Calculus (LCC) provides the newlineoperational semantics and the relevant methods needed in transforming NL statements newlineinto SQL, preserving the integrity and compositionality24 of expressions in every step of the translation. Though Lambda Calculus is found to be effective in representing the intermediate expressions and assists in performing the transformations that are needed for translating specific predicates into SQL, its inflexibility in combining parallel computations is a constraint. To represent clauses that are in parallel or are in pipeline, and to perform the required transformations on the intermediate expressions involving these, more advanced programming constructs are needed. It also adopts functional programming techniques to deal with complex scenarios involving nested queries. -
-
A Mixed methods study of psyhosocial factors in career decision making in adolescents
Career choice is an important developmental task in adolescence and is influenced by many factors. Using a mixed methods research design, this study aimed to understand career decision making and factors influencing the same in adolescents. In the quantitative phase the relationship between career maturity and perceived parenting style, personality traits, metacognition, socio- economic status, gender, college type, stream of study and decision status was studied in students studying in II Year Pre- University in Bangalore, India. Career decisions, personal and family factors in career decision making were explored in the qualitative phase. Informed consent was obtained from the participants and parents of the participants of the study. newlineQuantitative data was collected from 548 students studying in Arts, Science and Commerce stream in second year Pre- University in Bangalore. Students from eight private and seven government colleges were recruited for the study. Quantitative data was collected using a socio- demographic data sheet, Career Maturity Inventory, Parental Authority Questionnaire, Neo Five Factor Inventory and Metacognitive Awareness Inventory. The scales were translated to Kannada and back translated. In the qualitative phase, data was collected through a semi- structured interview schedule designed for this study. 30 students who were a part of the quantitative phase took part in this phase. The interviews were audio-recorded and transcribed for analysis. Statistical analysis was done to analyze quantitative data. Descriptive statistics, correlation, regression analysis, t tests and one-way ANOVA was done. Qualitative data was analyzed by template analysis and themes were derived from the data. The results revealed associations between personality traits neuroticism, openness and conscientiousness and specific aspects of career maturity attitude and competence. -
A mixed methods study on factors associated with relapse of alcohol use disorder
Background: Alcohol use disorder (AUD) is one of the most concerning mental health issues in India. According to the recent survey, Magnitude of Substance Use in India, 2019, 160 million of the countrys population consumes alcohol. About 35.6% are problem drinkers among those who drink, of which 18% are alcohol dependent. Despite the greater understanding of alcohol use disorder (AUD) and the scientific advancements in treatment, relapse remains to be the main challenge in managing AUD. This study aimed at investigating various factors associated with relapse of AUD and presenting an in-depth understanding of it. Methods: A Sequential Explanatory Mixed Methods design was used. In the quantitative phase, 72 relapsed individuals with AUD currently undergoing treatment were compared with 72 individuals previously treated for AUD who maintain total abstinence for a minimum period of one year. Relapsed participants were selected from three private de-addiction centers in Bangalore and abstaining participants were recruited from various Alcoholics Anonymous meetings in Bangalore. The relapsed and sober groups were matched on gender, AUD diagnosis, and previous inpatient alcohol de-addiction treatment. Cloninger's Temperament and Character Inventory-Revised was used to assess the personality profiles of the participants. A sociodemographic and clinical information form was also used to collect data. Six participants were selected purposively from the same sample for in-depth interviews. Data analysis was conducted using SPSS and NVivo for quantitative and qualitative data, respectively. The study protocol was approved by the institutional ethics committee. Results: Bivariate analyses showed a significant difference in Novelty Seeking, Persistence, Self-Directedness, and Self-Transcendence traits between the relapsed and sober participants. Results also suggested that reported use of other substances, post- discharge follow-ups, and living with drinking or drug-using individuals are significantly associated with relapse. Logistic regression displayed incomplete treatment, use of other substances, and no post-discharge follow-up as predictors of relapse. The qualitative thematic analysis revealed preparedness, motivation, personal exceptionalism, meaning and purpose, and social and interpersonal as the main relapse-related themes. Conclusions: The findings highlight the importance of treatment engagement, discharge planning, aftercare, and special attention to those presenting with multiple substance use. It also displays a few culture-specific aspects to be considered during treatment, such as preparing the individuals entering treatment to effectively engage, assessing and working with their motivation, and addressing the relationship dynamics. -
-
A Model for Churn Prediction Based on Qualitative Support Interaction Features for Hotel Technology Provider
Customer retention is a significant driver of a company s growth. Machine learning has gained immense popularity as a means to predict customers at risk of churn. Churn prediction models are capable of highlighting customers who are at high risk of churn well in advance. A popular approach to improve the performance of churn prediction models is by using input variables that are mainly quantitative and structured in nature. There are limited works in literature that newlineinvestigate smart means to effectively utilize and integrate unstructured data into churn prediction models, and study the impact on model efficacy. One of the roadblocks to effectively utilize unstructured data is the associated cost of annotation which is both time consuming and requires intensive manual effort. To overcome this obstacle, researchers often adopt a semi-supervised newlineapproach called active learning that aims to achieve state-of-the-art performance using minimal number of samples. Although active learning boosts classifier performance, the underlying query strategies are unable to eliminate redundancy in selected samples for manual annotation. Redundant samples lead to increased cost and sub-optimal performance of learner. Inspired by this challenge, the study proposes a new representation-based query strategy that selects highly newlineinformative and representative subsets of samples for manual annotation. Data comprises newlinemessages of a set of customers sent to a service provider. Series of experiments are conducted to analyse the effectiveness of the proposed query strategy, called Entropy-based Min Max Similarity (E-MMSIM), in the context of topic classification for churn prediction. The foundation of E-MMSIM is an algorithm that is popularly used to sequence proteins in protein databases. The algorithm is modified and utilized to select the most representative and informative samples. The performance is evaluated using F1-score, AUC and accuracy. -
A Model for storage optimization of brain MRI images for tumor detection using image processing technique
Magnetic Resonance Imaging (MRI) is a major non-invasive method for Brain tumor detection. The anatomical assessment of brain newlinetumor can be carried out using brain MRI image analysis. MRI is widely used in brain tumor identification and classification. Images generated during the diagnosis purpose are unattended after the specified diagnosis. newlineBrain MRI images in Digital Imaging and Communications in Medicine (DICOM) format require large amount of storage space. newlineAccumulation of the MRI images put forward the requirement of more storage space. To store large number of images, existing storage models has to be handled wisely. Research associated with storage, process and newlinemanipulation of medical image data using modern technologies with a minimal human intervention is the need of the time. Image processing deals with the study and development of innovative technologies for newlineanalysis,representation and interpretation of the image data. In this research, the need of an efficient storage model that can help in storing the brain MRI images is studied with the help of image processing technique. To store the brain MRI images with a reduced storage space, a matrix-based method is proposed. In this model brain MRI images in newlineDICOM format are converted into matrix format. In the DICOM images, the image data and header information together hold the details of the patient and image data. These data are converted and stored in the matrix. newlineThe stored matrix is accessed as the input to the proposed model. The proposed model follows different image processing steps.The process starts with pre-processing of brain MRI images followed by clustering of newlinewhite matter (WM), gray matter (GM) and cerebrospinal fluid (CSF), segmentation of tumor and classification of tumor and finally it handles the storage of the MRI images. In the pre-processing step, filtering algorithms are applied on MRI to remove the noise and text artifacts. The newlinewhite matter, gray matter and CSF are separated using the K-means newlineclustering method. -
A Model for the Secured Data Transfer of Healthcare Data by Image Steganography and Cryptography Techniques
Health care domain and related issues have evoked a lot of attention from researchers in the recent past. Ensuring the Security and Privacy of data of patients having classified disease is of utmost importance to the health care sector. From time immemorial cryptography and steganography techniques were used to provide data security. In health care domain, classified disease is the area of investigation as the patients having classified disease always prefer more security and privacy. The objective of this research work is to extract the benefits of cryptography and steganography techniques and to apply the combination of these security mechanisms to develop an algorithm that gives more security and privacy than existing techniques. The proposed algorithm builds on Rivest, Shamir and Adleman (RSA)algorithm which is considered to be one of the classical algorithms in cryptography. RSA is widely used in the encryption-decryption process for data transfer in internet which ensures the security of data in transit. The data encrypted using RSA is provided with one more layer of protection by calculating the message digest of the same. The message digest of encrypted data is calculated using message digest (MD5) algorithm. The steganographic approach of spread spectrum gives better security as it is always robust against statistical attacks and provides added security to the cryptographic protected data. Finally the Discrete wavelet transform (DWT) method is used for transformation. Combination of cryptographic technique and steganographic techniques offers higher level security to the data dealt in health care domain. Image type is a major factor contributing to the performance of the proposed algorithm. In recent years the open sources such as internet shows rapid usage of images especially Joint Photographic Experts Group (JPEG) image format as it occupies comparatively less space and provides better image quality even after applying image transformation. So medical image format type selected for the discussed research work is of JPEG format. Medical images of patients like X-ray of lung, bilateral section of face etc. are taken as cover image for the developed algorithm. Patients personal data is very confidential which is in the textual format is encrypted using RSA algorithm, then message digest is incorporated to keep the encrypted data homogeneous and then it is embedded in the cover image using DWT technique. The performance of the algorithm developed is measured on the basis of peak signal to noise which is a statistical method. The peak signal to noise ratio measurement is used as a visual quality measurement to simulate human perception as the difference between original image and embedded image seems to be identical as per developed algorithm. PSNR ratio is measured for various medical images selected. Noise ratio is also measured for Lena image as it is considered as a standard cover image in the steganography. Noise ratio of the selected medical images are measured and compared against cropped medical images. The results of the above research exercise presented in the thesis, infers that the proposed algorithm Raster Data protection algorithm provides enhanced security to the embedded medical images dealt within the health care sector, with minimal side effect on the quality of the image. KEYWORDS: Cryptography, Steganography, Health care, Security, JPEG, PSNR. -
-
A Novel Approach for Sensitive Crop Disease Prediction Based on Computer Vision Techniques
Agriculture is a vital sector that plays an essential role in ensuring global food security, supporting economic development, and promoting environmental sustainability. Sustainable agriculture is an essential approach that aims to address the diffculties posed by conventional farming practices and ensure the long-term viability of our food production systems. Worldwide, crop leaf diseases seriously threaten food security and agricultural production. Early and accurate detection of crop leaf diseases is essential for effective crop productivity management and food prevention. Computer vision approaches offer promising solutions for automating the identifcation and prediction of crop leaf diseases. Analyzing digital images of plant leaves enables the identifcation of disease characteristics, such as discoloration, lesions, and patterns, which are often imperceptible to the naked eye. Machine Learning (ML) algorithms, such as Convolutional Neural Networks (CNN), have been widely employed in this domain to learn from large datasets of annotated images and accurately classify leaf diseases. The process of crop leaf disease classifcation using computer vision involves several stages. Initially, highresolution images of plant leaves are acquired using cameras or mobile devices. Preprocessing techniques, including image enhancement and noise reduction, are applied to improve image quality. Subsequently, feature extraction approaches extract pertinent data from the images, including texture, shape, and color. Deep Learning (DL) models are then trained and fne-tuned using these extracted features. newlineAlthough computer vision techniques have shown effective results in the classifcation of plant diseases, however, several challenges remain. Tomatoes and Potatoes newlineare widely cultivated and consumed vegetables worldwide and are a primary economic newlinesource for many countries. These sensitive plants are prone to various diseases during newlinegrowth, leading to signifcant losses in productivity and fnancial impact on farmers. -
A novel framework for cloud-based analytic of massive and multi-structured healthcare images for real-time insights
The dependency of healthcare industry on the information and communication technology newline(ICT) domain is consistently on the rise in order to conceptualize and provide1537608 newlinesophisticated services to various newlinestakeholders including patients, newlinecaregivers, support service providers, medical practitioners, and experts. There are a variety of decisive advancements in the diagnosis, medication and surgical processes, medical electronics, instruments and equipment, healthcare-centric robots, a bevy of cloud-based healthcare software solutions, medical data hubs, etc. One direct offshoot of all these developments is that the amount of multi-structured data is exponentially growing. There is a litany of support and expert systems in order to lessen the doctors workloads. However the brewing challenges and new-generation requirements include the real-time processing of medical data to extract real-time insights and decision-enablement, the substantial enhancements in appropriate and accurate processing and understanding of various and overlapped symptoms towards correct and strategically sound decisions, the real-time analytics of medical data, the empowerment of medical devices to assist surgeons and specialists in performing their tasks in an assured manner, etc. newlineThe Problem Description - Medical imaging is one of the fundamental and most important areas of the healthcare system. This needs accuracy in processing and producing best results for further diagnosis and action. There are various factors impelling medical imaging like patient preparation, different scanning modalities, the scanner used to capture the image and various algorithms adopted for processing the captured images. -
A perspective analysis of emotional appeal used in television advertising /
The purpose of the study is to find out whether emotional appeal is still prevailing in television advertising. The researcher focuses on the various elements used by advertisers to evoke emotional response on the audience‟s side. The advertisements decided by the researcher portray important relationships that are valued and maintained in the society. -
A perspective analysis on doodle art used in five educational materials /
Doodle Art are simple drawings that have concrete representational meaning in abstract shapes. A doodler intently shifts through information to generate substantial understandings. Doodle Art is one of the evolving styles that are attracting young audience especially through subject materials. Themain aim ofthe study is to understand whether doodling has emerged as a new trend in brand recall. -
A perspective reading of photographs of the Ahmedabad city /
The present study talks about the usage of photographs as a major tool for storytelling. It shows how various photographs of various places, if shown to people and analyzed according to their perspectives and experiences can give us a rich definition and an idea about the events that had shaped the place and led to their current being. Also, it describes the place according to the way or the angle the respondents saw it to be. -
A Phenomenological Investigation of Mens Experiences of Depression and Gender Role Socialization in Early Family Relationships in urban South India
With the overwhelming suicide rate among men over women, and family problems and illness being the major causes of suicide in India, it is vitally important that we have a better understanding of how men struggle with depression in the context of family. Based on increasing studies on depression and family in the last two decades, a number of researchers have suggested that male depression is associated with early family life experiences. As few studies on the experiences of depression and gender role socialization in early family relationships are reported in India, the present study adds credence to the concept of male depression that may relate to early family life experiences. The aim of this study is to understand the subjective experiences of depression and gender role socialization in early family relationships among Indian men. A non-clinical sample of 9 men was selected using purposive sampling from a human service organization. Theoretical sampling of biographical accounts of a clinical group (1 male client with history of clinical depression) was used for triangulation of data. Consensual Qualitative Research methodology is adopted. The subjective experiences of participants are examined by open-ended, in-depth interviews. The interviews are taped and transcribed. Identification of domains and core ideas, and cross-analyses are conducted on transcripts. A research team of three and an external auditor are employed for data analysis. The findings have contributed to new understandings of depression in the light of gender role socialization in early family relationships among Indian men. Implications suggest further studies of male depression in the family context, the challenge of family life education in India, and the importance of gender sensitivity in counseling with men. Keywords: men, depression, gender role socialization, early family relationships, consensual qualitative research. -
A Posthuman Analysis of Human - Machine Relationship in Select American Science Fiction Films
The research A Posthuman Analysis of Human Machine Relationship in Select American Science Fiction Films attempts to foreground the emerging posthuman scenario brought about by the explosion of Artificial Intelligence (AI) in contemporary life by analysing the posthuman representations achieved by depicting AI characters and their relationship with humans in the select American science fiction films. The primary texts for the study are Stephen Spielberg s AI: Artificial Intelligence (2001), Spike Jonze s Her (2013), Mathew Leutwyler s Uncanny (2015), and Drake Doremus Zoe (2018). The research analyses the posthuman newlinerepresentations in the select films using the methodological framework of philosophical posthumanism of Francesca Ferrando with its constituent elements of post-humanism, post-anthropocentrism, and post-dualism. The term posthuman in philosophical posthumanism refers to the critique of the notion of human preserved by the Western humanistic traditions. The three constitutive elements of philosophical posthumanism, namely, post-humanism, postanthropocentrism, and post-dualism, offer a revisit of the notion of human propagated by Western humanistic traditions and offer a renewed worldview of being human in the contemporary technocentric society where nonhuman agency is being widely newlinerecognized. From an epistemological perspective, this research adds to the evolving posthuman discussions, providing a new dimension to what it means to be a human and challenging the age-old assumptions about the human condition. -
A prospective study on portrayal of rural theme in selected Tamil films /
Tamil film industry is popularly known as Kollywood. It has its own unique elements which makes it different from other film industry in Indian cinema. The researcher has tried to find out if there is any kind of rural element in the films which she has chosen in chronological manner. The researcher has analyzed based on certain concrete parameters. -
A psychoanalytical study of surrealist elements in films /
Psychoanalysis has over the years been a centre of attraction and intense research and study by critiques, psychologists, sociologists, etc. due to its unique outlook at the world, Freud’s psychoanalytical theories have found their way onto films, which use the creativity of visuals, sounds, effects etc to create the world Freud claims to be hidden behind the human consciousness. No film can escape psychoanalysis as it frames the underlying reasoning behind human behavior and thus this forms a most intriguing realm of study. -
A psychoanalytical study of the nature and appearance of character analysing 6 songs each of Jim Morrison, Kurt Cobain and Janis Joplin (members of the famous club 27) /
To understand how character could be understood by analysing the greatest works of the above artists. This research is to find a possible connection to show that the songs written by the artist is not just a form of expression but also to show the world who they are, and their ideologies and most importantly what they feel about themselves.