Patent Number: 202241057386, Application: Kapil Tiwari.
The present invention applies to probabilistic inference employing secure multi-party computation (SMPC) methodologies including confidential training, inference, and model protection for vertically partitioned datasets. The proposed invention addresses privacy
complexities of vertically partitioned data owners, model owners, and ML clients using confidential training, inference, and model protection (CoTraIn-VPD), a technique to boost data, model parameter, and client input privacy during ML inference without affecting accuracy or performance.
Patent Number: 202241057386, Application: Kapil Tiwari.
The present invention applies to probabilistic inference employing secure multi-party computation (SMPC) methodologies including confidential training, inference, and model protection for vertically partitioned datasets. The proposed invention addresses privacy complexities of vertically partitioned data owners, model owners, and ML clients using confidential training, inference, and model protection (CoTraIn-VPD), a technique to boost data, model parameter, and client input privacy during ML inference without affecting accuracy or performance.
Patent Number: 202221053302, Applicant: Raghvendra Omprakash Singh.
IoT and deep learning based Emotion and hate speech recognition to examine the person mind state in home/health care ABSTRACT There are several applications for AEE, which stands for automatic emotion recognition. Various industries, including advertising, technology, and human-robot interactions, use the emotional responses of individuals as a signal. This paper analyses all pertinent scientific literature to determine the application of sensors. In these publications, numerous strategies that have already been researched or implemented are discussed.
Patent Number: 202241057163, Applicant: Elizabeth Susan John.
Previously, the majority of skin-applied treatments and goods based on food pulp were marketed in jars or tubes. Later, more items were kept in user-friendly and specialised containers; one of the most popular ones being the dispenser bottle type. Dispensing bottles are incredibly well-liked in the present market. Several items have been offered for sale in the market on dispenser types, particularly sanitizers gel, creams, shampoos, etc. However, even modern dispensers are unable to dispense all of the contents inside due to their design features and structural profile.
Patent Number: 202241053663, Applicant: Joel Sebastian.
A high percentage of elderly users depend on walking slicks to get better support or assistance in maintaining a good posture while walking. An ergonomically well-designed walking stick can only provide good postural stability. For the past few decades in India, no significant changes or innovations have been made to the conventional walking stick. The study identified that majority of the users are facing difficulty in handling these conventional walking sticks.
Patent Number: 202241051890, Applicant: Moloy Biswas.
As time changed audio technology rapidly developed at a fast pace. Even with the advancing technology, all portable earphones seem to offer a specified feature. Customers have to invest in different types of earphones and headphones based on the device setting and features they require. Our product Neo-X headphones bring different features for different occasions or contexts. Whether in a gymnasium, hiking, bicycling or listening to music on a flight, these earphones provide a premium sound quality. Currently, there are a wide variety of earphones and headphones available in the market.
Patent Number: 202231033186, Applicant: Shruti Mohapatra.
The present invention discloses a system that is configured to receive from a user an income range and a geographical scope of interest in a predefined set of income distribution and aggregating prediction data from the income range and geographical scope from one or more of the asset classes selected from the group consisting of currency, bond, commodity, and stock. Further, automatically determining changes over time within the income range of one or more statistical analysis values of the aggregated prediction data selected from the group consisting of mean, median, mode, difference of means, standard deviation, variance, tolerance levels, skewness, kurtosis, inflection points and Bayesian analysis.
Patent Number: 202231033186, Applicant: Shruti Mohapatra.
The present invention discloses a system that is configured to receive from a user an income range and a geographical scope of interest in a predefined set of income distribution and aggregating prediction data from the income range and geographical scope from one or more of the asset classes selected from the group consisting of currency, bond, commodity, and stock. Further, automatically determining changes over time within the income range of one or more statistical analysis values of the aggregated prediction data selected from the group consisting of mean, median, mode, difference of means, standard deviation, variance, tolerance levels, skewness, kurtosis, inflection points and Bayesian analysis.
Patent Number: 202241059619, Applicant: Dr.C.Rohith Bhat.
Signals received by one or more sensors and processed by a processor can contain features such as one or more audio, visual, linguistic, and physical features. One or more machine learning algorithms may be used to assess the audio, visual, linguistic, and bodily data, and the results may be used to infer an individual's emotional state. The most recent multimodal speech emotion identification techniques, with an emphasis on audio, text, and visual data. We offer a fresh, descriptive classification of approaches based on how they manage the dynamics of intra- and intramodality in the time dimension.
Patent Number: 202241053590, Applicant: Dr. S Balamurugan.
It is reported in literature that nearly 4.57 billion people access Internet, covering nearly 59% of global population as per 2020 statistics. With huge number of Internet users and large volumes of data, the need for secure and fault-tolerant web applications increases. Huge volumes of data are not only consumed, but are also converted and copied among multiple computing resources. Proposed is a Supervised Machine Learning Technique for efficient management of cloud resources.
Patent Number: 202241053590, Applicant: Dr. S Balamurugan.
It is reported in literature that nearly 4.57 billion people access Internet, covering nearly 59% of global population as per 2020 statistics. With huge number of Internet users and large volumes of data, the need for secure and fault-tolerant web applications increases. Huge volumes of data are not only consumed, but are also converted and copied among multiple computing resources. Proposed is a Supervised Machine Learning Technique for efficient management of cloud resources.
