Sap Se

Jayananda Appanna Kotri

Tarun Sharma

Sharad Kejriwal

Abinaya S

Yashwanth Dasari

Kavitha Krishnan

Manoj Kumar

Rohit Jalagadugula

Gopi Kishan

Natesan Sivagnanam

Diwakar Maurya

Amey Haram

Prateek Garg

Sindhu Gangadharan

Kalpana Karunamurthi

Maheswar Reddy Pulla

Shweta Mohanty

Techniques and solutions are provided for facilitating the definition and use of virtual environments. Typically, virtual environments require manual linking of virtual environment components, such as locations, avatars to be displayed, avatar interactions, and actions to be taken upon a selected interaction. This is in addition to actions to define, for example, a visual representation of a virtual location or an avatar. Disclosed techniques provide standardized data objects that can represent components of a virtual environment and relationships between such components. For example, aspects of a virtual environment can be predefined, and a user can configure the virtual environment with particular avatars and particular avatar content (such as media to be rendered for that avatar). In particular aspects, components of a virtual environment, such as avatars, can be determined at runtime, included based on attributes associated with a particular user for whom the virtual environment is to be rendered.

Configurable virtual environment definitions

In an example embodiment, factual question generation from freeform content is achieved through semantic role labeling and recurrent neural networks (RNNs). Specifically, semantic role labeling is used to identify an answer phrase so that it can be replaced with an appropriate question word. RNNs are then used to extract triples (Subject-Object-Predicate) from the sentence, and each of these triples can be used as an answer phrase/word. An RNN is then fed with training data to generate the questions more efficiently.

Automated question generation using semantics and deep learning

Methods and apparatus, including computer program products, are provided for a contextualized bot framework. In some implementations, there may be provided a method which may include receiving, by a bot of a contextualized bot network and via a client interface associated with the contextualized bot network, a request, the request comprising a text string sent by the client interface; generating, by the bot, a token based on the text string of the request; determining, by the bot, whether the token generated based on the text string of the request matches a stored token that is stored in a database, based on one or more parameters, the stored token being mapped to an address stored in the database, the address comprising instructions for responding to the request; selecting, by the bot, the address in response to the match between the token generated based on the text string of the request and the stored token to obtain data at the address; and presenting, by the bot, at the client interface, data obtained at the selected address to form a response to the request. Related systems, methods, and articles of manufacture are also disclosed.

Contextualized chat bot framework

A recurrent neural network (RNN) is trained to identify split positions in long content, wherein each split position is a position at which the theme of the long content changes. Each sentence in the long content is converted to a vector that corresponds to the meaning of the sentence. The sentence vectors are used as inputs to the RNN. The high-probability split points determined by the RNN may be combined with contextual cues to determine the actual split point to use. The split points are used to generate thematic segments of the long content. The multiple thematic segments may be presented to a user along with a topic label for each thematic segment. Each topic label may be generated based on the words contained in the corresponding thematic segment.

Thematic segmentation of long content using deep learning and contextual cues

Growing community of inventors

Bengaluru, India

Jayananda Appanna Kotri