Telefonaktiebolaget Lm Ericsson (publ)

Adobe, Inc.

Tata Consultancy Services Limited

Balaraman Ravindran

Subramanian Shivashankar

Balaji Vasan Srinivasan

Ritwik Sinha

Anandhavelu Natarajan

Avinash Achar

Ankur Dauneria

Vineet Gupta

Shriram Revankar

Saket Gurukar

Gokul Karthik

Sayan Ranu

Vishnu Sankar

This disclosure relates generally to method and system for forecasting sales based on N-Gram model. The present disclosure provides accurate prediction of sales for optimal operations to reduce the cost. The method receives a plurality of inputs of each product comprising a sales history, and a current price bin. The categorical sale(s) for each product is discretized based on the sales history by clustering each product sales history into a one or more groups based on a maximum sales velocity range. Further, a probability table is generated for the discretized categorical sales of each product based on computing a round off weighted mean and a median using a N-Gram model. Then, a smooth probability table is computed for the generated probability table. To forecast sales multistep prediction for the smooth probability table is computed based on at least one of a joint approach, a bootstrapped approach, and a step greedy approach.

Method and system for forecasting sales based on N-Gram model

The present disclosure relates to a method performed by a network node () in a communication network () comprising a plurality of communication devices (), for finding temporally connected connection patterns of the communication devices in the network. The method comprises identifying signalling between the communication devices during a predefined time duration to form a main communication graph in which the plurality of communication devices are nodes. The method also comprises partitioning the main communication graph for a time period comprised in said time duration to capture temporally connected signalling between some of the communication devices as illustrated in the main communication graph in a partitioned communication graph. The method also comprises forming at least one subgraph from the partitioned communication graph, comprising m nodes corresponding to m communication devices of said some of the communication devices. The method also comprises constructing a binary vector of the subgraph over the time period divided into a plurality of sampling periods corresponding to components in the vector such that, for each component in the vector, the component value is set to '1' if, in accordance with the subgraph, signalling occurs between at least two of the m communication devices during the sampling period corresponding to the element, and the component value is otherwise set to '0'. The method also comprises testing whether the signalling between the different m communication devices used for constructing the binary vector is temporally connected by checking that the number of consecutive '0' element(s), if any, between any two &#x201c;1&#x201d; elements is less than a predefined &#x394;T.

Analysis of connection patterns in a communication network

A method for detecting communities in a network comprises the steps of defining an internal scoring function (IS) for a community, wherein the internal scoring function quantifies an internal quality of the community, and defining an external scoring function (ES) for a community, wherein the external scoring function quantifies an external quality of the community. A community scoring function (IE) for a community is determined, wherein the community scoring function comprises a product of the internal scoring function and the external scoring function. A network scoring function is determined, wherein the network scoring function is based on the community scoring functions of communities in the network. The network scoring function is used by a maximization algorithm, for example a greedy maximization algorithm, for detecting a community in a network.

Method and apparatus for detecting communities in a network

This document describes techniques for predicting reactions to short-text posts. In one or more implementations, a prediction model for short-text posts is generated from previous posts to a social network and responses to the posts by the social network community. Subsequently, the prediction model can be used to predict the social network community's reaction to a proposed post prior to the proposed post being posted to the social network.

Predicting reactions to short-text posts

Growing community of inventors

Chennai, India

Balaraman Ravindran