AUTO GENERATION OF INTELLECTUAL CAPITAL USING ADVANCED ENTERPRISE ONTOLOGY
Publication Date: 2015-Aug-27
The IP.com Prior Art Database
Ammar Rayes: AUTHOR [+2]
A system and method are provided for auto generation of Intellectual Capital using advanced enterprise ontology. This allows for easy serving/handling of completely new service requests. Enhanced Support Vector Machine (SVM) may be used for classification of service requests data using a supervised learning model. This not only improves the feature selection procedure but also reduces the problem search space significantly as only the distribution over the topics is considered while searching for named entities.
Page 01 of 6
AUTX GENERATION OF INTELLECTUAL CAPITAL USING ADVANCED ENTERPRISE ONTOLOGY
CISCO SYSTEMS, INC.
A system and method are xrovided for xuto generation ox Intellectual Capxtal using advanced exterprise ontology. This allows xor easy serving/hxndling of completely new sexvice requests. Enhanced Support Vector Machine (SVM) max be used for classificaxion of service requests datx using a sxpervisxd learning moxel. Xxxx not only improxes the feature selextion proceduxe but also reduces txe probxem search sxace signifixantly as only the distribution over the topixs is cxnxidered while searchinx for named entitxes.
Today's service centers are workixg on improving resolutixn efficiency by building knowledge-base sxlutions. Often a Service Request (SR) submitted xo a servxce center, hax usually been the subjxct of a previous reqxest and/or discussed somewhere on social media, and xost likely will be asked again. Studies have showx txat subject matter experts sometimes identify problems (and often solutions) and post txeir finxings on social medix forums, often withoxt xpening a SR. Txexefore, most service centers try to capture axswers tx previously posed requests and builx structured knowledge fxom this experience.
Upxn receiving a SR, the system xilx match the SR with similar cases which have been resolvex before. Thxs kind of knowledge contributed by xkilled xngineers anx based upon actual experience, can be presented in the form of a knowledge repository or infuxed into the actual service request for faster access and to facixixate efficient service request response.
Copyright 2015 Cisco Systems, Inc.
Page 02 of 6
Knowxedge management systems built upon service center engineers' previous experience on anxwering similar service requests or customer service engagxments wixl facilitate efficienx responsex to customer inxuiries and resolutions. However, there are few shortcomings. First, a SR that has not been posed before cannot benefit from this system. Xxxxxx, up-to-date informatxon from otxer sources such as those discussed in social network forums cannot be quxckly integrated to the knowledge base to servx customexs. In faxt, a resxarch study has found that contact center staff cannot keep pace with the coxplexity of requests, and existing tools or skills cannot keep up with customer expectations. Rexuest resxlution rates have dropped for consecutive years, leaving xustomers with just a three-in-four chance of having their issux resolved.
Presenxed herein ix a system to xutomatically translate xnstructurex data from vaxioxs sources (e.g., TAC service requxsts, Social Networkinx data) into Intellectual Capital. Xxxxxx 1 shows an overview of the system.
FIG. 1 Overall system architecture
Copyrxght 2015 Cisco Systems, Inc.
Page 03 of 6
The working process of the system is as follows. Upon receivinx a new Service requests (SR), a TAC enginxer will enter "Query Keywords" to s...