International Business Machines Corporation

Taiga Nakamura

Aly Megahed

Hironori Takeuchi

Kugamoorthy Gajananan

Peifeng Yin

Mark A Smith

Mari Abe Fukuda

Futoshi Iwama

Hamid Reza Motahari Nezhad

Pawan Raghunath Chowdhary

Heiko Hary Ludwig

Robert John Moore

Shun Jiang

Nathalie Baracaldo Angel

Shubhi Asthana

Ken Mizuno

Shuichi Shimizu

Valeria Becker

Seiji Kobayashi

Mu Qiao

Ryuki Tachibana

Anbang Xu

Zhe Liu

Hiroaki Nakamura

Hiroshi Horii

Yunyao Li

Rama Kalyani T Akkiraju

Akira Koseki

Tamiya Onodera

Hovey R Strong, Jr

Hovey Raymond Strong

Mark Andrew Smith

Hiroshi Maruyama

Takayuki Kushida

Debdoot Mukherjee

Karen Frida Yorav

Shuhichi Shimizu

Scott Ross Trent

Michael E Rose

Brian D Yost

Hyun-Woo Kim

Tamotsu Sogo

A computer-implemented method according to one embodiment includes receiving, at a scheduler, a training data instance and a target instance, generating, by the scheduler, an input sequence from the training data instance and the target instance, sending the input sequence from the scheduler to an encoder, mapping, by the encoder, the input sequence to a feature vector, sending the feature vector from the encoder to the scheduler, sending the feature vector from the scheduler to a predictor, and mapping, by the predictor, the feature vector to a class vector to create a label for the target instance.

Predicting functional tags of semantic role labeling

One embodiment provides automatically learning shared resource environment solution design rules from a collection of requirement-solution pairs including obtaining requirement-solution pairs for a shared resource environment from a data store. A processor iteratively generates a candidate design rule set from each requirement-solution pair. Each generating iteration uses an input including the candidate design rule set output from a previous generating iteration. Evidence scores of each candidate design rule are calculated and candidate design rules having higher evidence score than an evidence score threshold are retained to obtain a learned design rule set. Candidate rules of a next iteration are constructed based on an addition of new attributes to rules of the learned design rule set.

Automatic rule learning in shared resource solution design

One embodiment provides a method that includes receiving adjusted labeled data based on emotional tone factors. Words are analyzed using a tone latent Dirichlet allocation (T-LDA) model that models tone intensity using the emotional tone factors and integrating the adjusted labeled data. Representative words are provided for each emotional tone factor based on using the T-LDA model. The representative words are obtained using the T-LDA model based on determining posterior probabilities and adjusting the posterior probabilities based on an auxiliary topic.

Machine learning from tone analysis in online customer service

A computer-implemented method, according to one embodiment, includes: receiving an offer request including one or more desired services, and selecting available offerings, each of which include at least one of the desired services. A determination is made whether available benchmarks exist for each of the at least one desired service included in each of the selected available offerings. For each desired service determined as not having available benchmarks, a draft benchmark is computed for each of a plurality of criteria. A confidence weight is also computed for each of the draft benchmarks. The available benchmarks, the draft benchmarks, and the confidence weights are further used to construct an offer which is submitted in response to the received offer request. Moreover, the draft benchmarks and the corresponding confidence weights are re-computed for each of the respective desired services in response to determining that the submitted offer was not accepted.

Benchmark scalability for services

One embodiment provides a method for augmenting missing values in historical or market data for deals. The method comprises receiving information relating to a set of deals. For any service included in one or more deals of the set of deals but not included in one or more other deals of the set of deals, the method further comprises augmenting, for any or all of the one or more other deals that does not include the service, one or more missing values for the service with one or more recommended values based on a recommendation algorithm. The service may be at any service level of a hierarchy of services.

Augmenting missing values in historical or market data for deals

One embodiment provides model-driven and automated generation of information technology (IT) solutions including obtaining a set of business and technical requirements for IT infrastructure and applications. A client business and technical requirement model is generated based on generic model constructs and extending with constructs specific to capturing client requirements. A draft IT solution is generated using an automated model-driven process to generate the draft IT solution configuration for client requirements for a target shared resource environment offering. The generated draft IT solution is translated into a language of a constraint satisfaction engine that propagates values of chosen attributes in the draft solution to identify valid values for unset attributes, and identifies conflicts. An IT solutions interface is generated based on auto-population of verified attribute results.

Model-driven and automated system for shared resource solution design

One embodiment provides a method for estimating unit price reduction of services in a new in-flight deal using data of historical deals and market reference deals cost structures. The method includes receiving a detailed cost structure for historical information, market deals information, services quantity information and deals metadata for a first year. For each service: peer deals to the in-flight deal are selected based on the detailed cost structure; missing cost data values in the peer deals are augmented; unit cost reduction values for the peer deals estimated; the unit cost reduction for the in-flight deal from each year in total contract years to a next year without a last contract year are estimated; and a total cost for the in-flight deal for all years in the total contract years beyond the first year are estimated.

Estimating annual cost reduction when pricing information technology (IT) service deals

One embodiment provides a method comprising receiving historic peer deals relating to at least one service, and a baseline and cost percentage estimation for each service. Historic peer cost data for each service is clustered to form at least one cluster. Each cluster includes similar unit costs, and has an assigned label. A classification model is trained based on each baseline received, each cost percentage estimation received, and each assigned label. For each assigned label, a corresponding probability distribution is computed based on the classification model. For each service of a new client solution, an assigned label for the service is predicted based on the classification model, and, based on a probability distribution corresponding to the assigned label predicted, transforming an initial range of historic peer cost data relating to the service into a narrower range for use in estimating a cost of the service with improved accuracy.

Effectiveness of service complexity configurations in top-down complex services design

One embodiment provides automatically learning shared resource environment solution design rules from a collection of requirement-solution pairs including obtaining requirement-solution pairs. A processor iteratively generates a candidate design rule set from each requirement-solution pair. Candidate design rules from the candidate rule set are filtered to obtain a learned design rule set. The learned design rule set is optimized based on merging design rules.

A system for cognitive assessment of the competitiveness of deals may include a memory having stored thereon historical deal information for historical deals with each historical deal including a historical deal component. A historical deal component may include a historical work scope and associated historical work pricing. The system may also include a processor cooperating with the memory and configured to compare current deal information with the historical deal information. The current deal information may include a current deal component that may include a current work scope and associated current work pricing. The processor may use machine learning to determine whether the current deal component is non-competitive based upon the historical deal information, and for each non-competitive current deal component generate an alternative current deal component. The alternative current deal component may have at least one of a different current work scope and different associated current work pricing.

System for cognitive assessment of transactions

One embodiment provides a method for top-down pricing of an in-flight deal. The method comprises receiving a first set of information relating to the in-flight deal. The in-flight deal comprises multiple services. The method comprises, for each service of the in-flight deal, selecting a corresponding set of peer deals from historical and market data based on the first set of information, and mining costs for the service from the corresponding set of peer deals. The method further comprises determining a set of price points for the in-flight deal based on each cost mined.

Top-down pricing of a complex service deal

One embodiment provides a method comprising answering one or more incoming phone calls received at one or more pre-specified phone numbers utilizing a bot. The bot is configured to engage in a conversation with a caller initiating an incoming phone call utilizing a voice recording that impersonates a human being. The method further comprises recording each conversation the bot engages in, and classifying each recorded conversation as one of poison data or truthful training data based on content of the recorded conversation and one or more learned detection models for detecting poisoned data.

Bot-based data collection for detecting phone solicitations

A method for selecting a set of information technology (IT) services peer deals to an in-flight deal for each first level service in the in-flight deal includes receiving a detailed cost structure for historical information, in-flight deals information, market deals information, services baselines and deals metadata, and multiple peer deals for selection. For historical information and market deals information, all missing baselines for all services in the in-flight deal and all missing unit cost for services are augmented at the first level service. The multiple peer deals are classified into different clusters at the first level service. A closest cluster to the in-flight deal at the first level service is selected. For each second level service of the in-flight deal the method: classifies the selected peer deals into different clusters. A predetermined number of peer deals that appear in a largest number of the selected clusters is selected.

Selecting peer deals for information technology (IT) service deals

Growing community of inventors

Sunnyvale, CA, United States of America

Taiga Nakamura