ABSTRACT. Facial emotion detection is a technique for identifying human emotions from facial expressions. Autism Spectrum Disorder is an advanced neurobehavioral disorder, autism is a vast field with many problems, including hyperactivity, sensory, speech, etc. Children with autism who have behavioral problems are the focus of this research. This research is on an automatic detection system that will detect the expressions of autistic children and generate a report explaining their behaviors during their therapy sessions. This automatic system is primarily focusing on three main behavioral issues: happy, neutral, and irritated. This research is done to provide facilities to the parents, therapists, and Autistic Rehabilitation Centers to keep records of their children’s and patients’ progress in the form of reports. This is carried out using machine learning and image processing methods. The autistic children's faces are used to extract the features using local binary patterns. The proposed system uses CNN, Haar Cascade object detection Algorithm, and TensorFlow. The classification of emotions and face detection is done using the CNN algorithm and Haar Cascade object detection algorithm. Helping Hands Rehabilitation Center, a Pakistan-based autism rehabilitation center has provided its utmost assistance to make this research possible.

ABSTRACT. Farming is Pakistan's most basic job and plays a crucial part in the economy. A substantial portion of the land is dedicated to agriculture growing in order to meet people's demands for food, raw material export and import. As a result, it is critical to increase agricultural production, which is farmers' major competition. Crop cultivation is influenced by environmental and soil characteristics, which farmers are unaware of. A crop suggestion system is being developed to aid farmers in resolving this challenge. The construction of a system that suggests the best crop based on environmental and soil variables is done using machine learning techniques. Farmers that cultivate the recommended crop produce more and of higher quality.

ABSTRACT. Climate change has emerged as one of the greatest challenges the humanity is facing today due to its significant impacts on the environmental sustainability and ecosystem. The urban centers are considered as the prime hubs for the human induced climate change consequences due to the larger industrial activities, landuse changes and urbanization, deforestation, and release of Greenhouse Gases and aerosols into the atmosphere which results in the occurrence of urban heat islands and warmer climate in the urban cities. This study investigates the past annual and seasonal temperature trends over the five major urban centers of Pakistan including Karachi, Lahore, Islamabad, Peshawar, and Quetta for the period of 1990-2020 using the Mann-Kendall test. The Sen’s slope estimator method was used to detect the magnitude of change and the serial correlation in the time series data was removed using the Pre-whitening method. The results obtained showed a warming trend of annual mean temperature at Karachi, Lahore, Islamabad, and Quetta stations, while a significant decreasing trend was found at Peshawar station at 95% confidence interval. It is therefore concluded that the land use changes, rapid urbanization and other anthropogenic activities are among the major causes for the warming temperatures in the major urban centers of the country.

ABSTRACT. Due to its adaptability in harsh situations and resistance to signal distortion, the applications for fiber optics are virtually limitless. Through wireless broadband, wireless technology has enabled access to the internet at fast speeds. Radio over fiber is the combination of microwave and optical networks are the potential solution for reducing equipment, operational, and capacity increase costs. It is the optimal method for merging a wireless and wired environment because it combines the most desirable features of the two most widespread communication technologies. The objective of this study is to compare the performance of ROF in wireless environments such as LTE and WiMAX to that of conventional optical fiber. Radio technology employing optical fiber as a transmission medium may be the answer for the next generation. The research is then employed to determine the viability of ROF as a backhaul solution for LTE and WiMAX. MATLAB is used to deploy an optical and ROF network to acquire simulation results that validate the theory.

ABSTRACT. Hardware electronic manufacturing companies of the world producing a range of hardware products as per the end user requirements related with the field of medicine, communications, instrumentation, automation, power electronics, power transmission, surveillance, IoT etc. However, before sending these products to the market for the commercial use; their quality, reliability and functional requirements are verified are validated through various tests. The tests are conducted by employing various test sites that are designed according to the types and functional requirements of the hardware electronic products. In this survey paper, a brief review of hardware electronic products and its manufacturing life cycle is presented. The paper also covers the survey of test sites and the automation techniques (using machine learning) of these electronic hardware products. Finally, it will be concluded mentioning certain prime limitations and constraints of these test sites.

ABSTRACT. A lot of research has been done over the last few years in the field of facial Emotion recognition (FER) and it is currently a hot topic in both academia and industry. A good way to perceive a user's thoughts and design a good to detect universal facial expressions system. The option to separate different universal facial expressions based on emotional classes makes it feasible to recognize distinct emotions of facial expressions based on the diversity among facial calculations and expressions. Face recognition technology is used extensively today, often in the medical industry, the attendance system, and so on. This work focuses on developing a Facial Expression Recognition (FER) system using CNN. This the model uses Haar cascade classifier for real-time universal facial expressions using a webcam. It was trained using the [FER2013] dataset available on Kaggle to detect facial expressions.

ABSTRACT. This paper presents the development of model of financial sustainability and its impact on SMEs performance in Malaysia. The data used to the model was collected from questionnaire survey on the SMEs companies in Malaysia. The respondents were the managers of company that were requested to gauge each of the financial sustainability factors using 7-points Likert scale that they perceived affecting the company performance. A total of 146 valid responses were used for this analysis. After the model was constructed, it was evaluated at the measurement component of the model where it involved examining the indicator reliability, convergent validity, and discriminant validity. Then at structural component, it involved checking the strength of the relationship, checking coefficient of determination, conduct predictive relevance of the model, and conduct hypotheses testing. The significant relationships are financial sustainability and performance toward SMEs companies. These outcomes are from actual perception from the respondents where the collected data is not strong enough to trigger the significant relationship of other constructs that had been hypothesised. The model can help to give better understand to parties that concerned the financial sustainability factors in SMEs.

ABSTRACT. Diabetes is a chronic illness or group of metabolic disorders in which a person has a sustained rise in blood glucose levels (BG) because of a lack of, or an inability of, cells to respond to insulin. These days, this illness is causing severe health issues and long-term obstacles. Massive quantities of highly confidential material are present in the healthcare sector, and they must be handled properly. Diabetes Mellitus (DM) is regarded as one of the worst diseases in the world. For such examination of diabetes, clinical specialists require a trustworthy framework of objectives [20]. The collection includes data on 768 patients and the nine distinctive traits that correlate to them. To estimate blood sugar levels, we applied one ML systems to the sample [11]. We use KNN machine learning approach because it gives the perfect estimation of the dataset. Different Machine Learning (ML) techniques may be used to evaluate data from various perspectives and condense it into valuable evidence. The KNN method is used in this study to extrapolate diabetes [11]. In this paper, I have proposed future prediction of the diabetes in the human body.

ABSTRACT. With the continuous development of information technology and medical data information, more and more clinicians recognize artificial intelligence or will completely change medical practice by using advance machine learning methods. The potential of using machine learning and predictive analysis to customize specific treatments for individuals is currently under research. Machine learning can learn a large number of medical data, explore the dependencies in data concentration, thereby forming a corresponding medical model; which can quickly and accurately predict new data, which is conducive to the early diagnosis of diseases, and assisted clinical decisions. Clinical medicine faces the status quo of the relatively shortage of medical resources, the identification and rapid diagnosis and treatment of emergency critically ill patients. In the era of big data, the clinical demand is guided by clinical needs, the smart medical care of the machine is a means of solving the key to solving the above problems.