Abstract: Dynamic content tags are generated as content is received by a dynamic content tagging system. A natural language processor (NLP) tokenizes the content and extracts contextual N-grams based on local or global context for the tokens in each document in the content. The contextual N-grams are used as input to a generative model that computes a weighted vector of likelihood values that each contextual N-gram corresponds to one of a set of unlabeled topics. A tag is generated for each unlabeled topic comprising the contextual N-gram having a highest likelihood to correspond to that unlabeled topic. Topic-based deep learning models having tag predictions below a threshold confidence level are retrained using the generated tags, and the retrained topic-based deep learning models dynamically tag the content.

Abstract: Dynamic content tags are generated as content is received by a dynamic content tagging system. A natural language processor (NLP) tokenizes the content and extracts contextual N-grams based on local or global context for the tokens in each document in the content. The contextual N-grams are used as input to a generative model that computes a weighted vector of likelihood values that each contextual N-gram corresponds to one of a set of unlabeled topics. A tag is generated for each unlabeled topic comprising the contextual N-gram having a highest likelihood to correspond to that unlabeled topic. Topic-based deep learning models having tag predictions below a threshold confidence level are retrained using the generated tags, and the retrained topic-based deep learning models dynamically tag the content.

Abstract: To identify a target engagement sequence with a highest likelihood of realizing an opportunity, a target engagement sequence generator uses models (artificial recurrent neural network (RNN) and a hidden Markov model (HMM)) trained with historical time series data for a particular combination of values for opportunity characteristics. The trained RNN identifies a sequence of personas for realizing the opportunity described by the opportunity characteristics values. Data from regression analysis indicates key individuals for realizing an opportunity within each organizational classification that occurred within the historical data. The HMM identifies the importance of each persona in the sequence of personas with communicates to the key individuals. The resulting sequence of individuals indicates an optimal sequence of individuals and order for contacting those individuals in order to realize an opportunity.

Abstract: To identify a target engagement sequence with a highest likelihood of realizing an opportunity, a target engagement sequence generator uses models (artificial recurrent neural network (RNN) and a hidden Markov model (HMM)) trained with historical time series data for a particular combination of values for opportunity characteristics. The trained RNN identifies a sequence of personas for realizing the opportunity described by the opportunity characteristics values. Data from regression analysis indicates key individuals for realizing an opportunity within each organizational classification that occurred within the historical data. The HMM identifies the importance of each persona in the sequence of personas with communicates to the key individuals. The resulting sequence of individuals indicates an optimal sequence of individuals and order for contacting those individuals in order to realize an opportunity.

Abstract: A system and method for assessing visual field of a subject eye for any blind spots. The method includes the steps of presenting a screen that is rendered as a grid, wherein points of the grids are spaced in units of visual angle. A visual focus point is presented on the screen on the screen and a visual stimulus moves on the screen consecutively while the subject eye focuses on the visual focus point and tracks the movement of the visual stimulus. When the visual stimulus enters a blind spot area in the visual field, the movement of the visual stimulus is perceived to be halted but is moving. Upon exiting the blind spot area, the halted visual stimulus is perceived to start moving again. Inputs from the user are taken when the above two events occur. Based on the inputs, the area of the blind spot is mapped.

Abstract: Dynamic content tags are generated as content is received by a dynamic content tagging system. A natural language processor (NLP) tokenizes the content and extracts contextual N-grams based on local or global context for the tokens in each document in the content. The contextual N-grams are used as input to a generative model that computes a weighted vector of likelihood values that each contextual N-gram corresponds to one of a set of unlabeled topics. A tag is generated for each unlabeled topic comprising the contextual N-gram having a highest likelihood to correspond to that unlabeled topic. Topic-based deep learning models having tag predictions below a threshold confidence level are retrained using the generated tags, and the retrained topic-based deep learning models dynamically tag the content.

Abstract: To identify a target engagement sequence with a highest likelihood of realizing an opportunity, a target engagement sequence generator uses models (artificial recurrent neural network (RNN) and a hidden Markov model (HMM)) trained with historical time series data for a particular combination of values for opportunity characteristics. The trained RNN identifies a sequence of personas for realizing the opportunity described by the opportunity characteristics values. Data from regression analysis indicates key individuals for realizing an opportunity within each organizational classification that occurred within the historical data. The HMM identifies the importance of each persona in the sequence of personas with communicates to the key individuals. The resulting sequence of individuals indicates an optimal sequence of individuals and order for contacting those individuals in order to realize an opportunity.

Abstract: To automatically identify a sequence of recommended account/product pairs with highest likelihood of becoming a realized opportunity, an account/product sequence recommender uses an account propensity (AP) model and a reinforcement learning (RL) model and target engagement sequence generators trained on historical time series data, firmographic data, and product data. The trained AP model assigns propensity values to each product corresponding to received account characteristics. The trained RL model generates an optimal sequence of products that maximizes the reward over future realized opportunities. The target engagement sequence generators create target engagement sequences corresponding to the optimal sequence of products. The recommender prunes the optimal sequence of products based on the propensity values from the trained AP model, the completeness of these target engagement sequences, and a desired product sequence length.

Abstract: The innovation involves the use of a laser to ablate a calculated microstructure or employ an adaptation of maskless photolithography using a Digital Micromirror Device to serve as a Spatial Light Modulator to embed a covert diffraction screen, holding encrypted information under transparent surfaces of plastics or glass substrates. One method includes the steps of fragmenting the calculated diffraction screen into at least first and second parts that are placed in separate locations. In this method the binary pattern in each of the parts includes information representing a respective portion of the original image and needs to be interrogated simultaneously to provide a meaningful visual output: each part by itself being incapable of generating any meaningful information. The fragmentation method allows a public-private key type of secure system platform.

Abstract: Methods and apparatus for making, encoding, and encrypting binary patterns for storing recorded information. One of the methods includes the steps of providing an original image, fragmenting the original image into at least two parts including a first part and a second part, forming a mirror image of one of the a least two parts, inverting the first and second parts, forming a binary pattern of the first part and the second part, and forming a mirror image of the binary pattern of the one of the a least two parts. In this method the binary pattern in each of the parts includes information representing a respective portion of the original image.

Abstract: Methods and apparatus for making, encoding, and encrypting binary patterns for storing recorded information. One of the methods includes the steps of providing an original image, fragmenting the original image into at least two parts including a first part and a second part, forming a mirror image of one of the a least two parts, inverting the first and second parts, forming a binary pattern of the first part and the second part, and forming a mirror image of the binary pattern of the one of the a least two parts. ,In this method the binary pattern in each of the parts includes information representing a respective portion of the original image.