A Preliminary Short Survey of State of the Art Enterprise Search

Engines for Future Work Place

Selver Softic, Manfred Rosenberger, Markus Zoier, Konstantin Mondelos

and Erik Pillinger

Virtual Vehicle Research and Test Center, Graz University of Technology, Inffeldgasse 21A, Graz, Austria

Keywords:

Survey, Evaluation, Enterprise Search, Future Work Place.

Abstract:

This short survey represents a ﬁrst step towards identifying relevant enterprise search engines as possible key

enablers for challenges related to Future Workplace like knowledge transfer and individual and organisational

information management inside an enterprise collaboration cycle. In the early period of our research we

want to deliver a short overview on current state of the art solutions that can contribute to the idea of Future

Workplace. We summarised in our literature study most signiﬁcant parameters from current research on the

topic of semantic and enterprise search. Information was collected via product sheets and white paper of the

vendors as well using previous studies and accessible information at the web. Four most advanced solutions

in this area has been evaluated. We used them to check their appliance for Future Workplace trends and are

aiming at expansion of our review for additional solutions in the future.

1 INTRODUCTION

The Future Workplace tries bringing answers as:

which tools, services, processes and actors will be

the most challenging for a Knowledge Worker in the

near future. One of them is the search and consum-

mation of relevant information in a more precise and

faceted way. Hereby the choice of adequate retrieval

technology has to sufﬁce to the demands like mobil-

ity, ﬂexibility, privacy issues as well as precision re-

garding search results. Search engines cannot suf-

ﬁce the information need and the speciﬁc demands

of the Knowledge Worker. This rises a requirement

for a new generation of search engine frameworks.

Especially in the enterprise surrounding inquiry for

such solutions is growing with the complexity of data

that should be managed for every day working pro-

cess. We analysed the needs identiﬁed by the other

studies in this ﬁeld and deﬁned upon these ﬁndings

a subset of eminent features as criteria for evaluating

of enterprise search systems. Since no other studies

treated enterprise search for Future Workplace we de-

cided to choose the most prominent provider in this

ﬁeld and to evaluate them, using the extracted fea-

ture set. Enterprise search represents a step forward

regarding the strong data cross-linking which is one

of the main demands for Future Workplace as current

research claims (Denger et al., 2012). In this short

survey we present results of our evaluation for prod-

ucts: Sinequa, Conveawer, Semaphore and IntraFind.

This short paper is organised as follow: After intro-

ductory note we will describe some related work re-

garding Knowledge Worker, Future Workplace, enter-

prise search studies which tried to deﬁne the key con-

cepts to evaluate such systems. Further we will move

on with precise description of our methodology and

describe the results of our evaluation made upon de-

rived feature set on four enterprise search engines. We

will ﬁnally draw conclusions from evaluation results

and describe the ongoing future work.

2 RELATED WORK

Knowledge Worker as such has been natively deﬁned

over 50 years ago by Drucker in (Drucker, 1959), as

a person who is payed for intellectual work and cor-

responding knowledge. In today’s enterprises the im-

portance of this role grows with the knowledge cumu-

lation through increased information ﬂow and cross

linking. Almost everybody involved in decision pro-

cesses acts as Knowledge Worker inside the collabo-

ration cycle. Future Workplace therefore as such de-

ﬁnes new demands on roles as such involved within

the whole process with the occurrence of technolo-

626

Softic S., Rosenberger M., Zoier M., Mondelos K. and Pillinger E..

A Preliminary Short Survey of State of the Art Enterprise Search Engines for Future Work Place.

DOI: 10.5220/0004505706260629

In Proceedings of the 9th International Conference on Web Information Systems and Technologies (FWP-2013), pages 626-629

ISBN: 978-989-8565-54-9

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

gies like search engines, social media, cloud com-

puting etc., as current research on this topic reveals

(Denger et al., 2012; Stocker et al., 2012). Further,

according to the recent works the need for increased

faceted information consumption with the high degree

of mobility and ﬂexibility regarding the service which

provides support for this demand is one of the key en-

ablers of Future Workplace. Such demands require

more then a simple keyword based search as provided

by state of the art search engines like Google

. These

demands therefore raised the question of enterprise

search engines.(Mukherjee and Mao, 2004) described

the main aspects in more generic way. Studies like

(Hager, 1997; Ulbrich et al., 2009; Gronau, 2005)

have shown that demands on retrieval, data cross-

linking and information consumption for an enter-

prise search system should provide overlap in a cer-

tain set of features. Based upon their ﬁndings and

demands related to the idea of Future Workplace we

summarized the most signiﬁcant aspects and gener-

ated evaluation feature set for the semantic enterprise

engines in our survey.

3 METHODOLOGY

In our current research we are seeking for solu-

tions that support management of structured, semi-

structured and unstructured data. Desired solution

should also at least include personalised search with

variety of options and visualisation of such results

along with navigation facets and search history sup-

port. Access rights management on a huge data pool

has been also deﬁned as one of the criteria relevant for

the choice of evaluation candidates. Sinequa

, Con-

Weaver

, Semaphore

and IntraFindIntraFind

rep-

resent this level of enterprise search solution which

has qualiﬁed them as candidates for evaluation.Our

methodology includes following steps: First, we iden-

tiﬁed a subset of studies that deal on feature detection

(Hager, 1997; Ulbrich et al., 2009; Gronau, 2005) for

rating the enterprise search solution. After comparing

the ﬁndings in this literature work we derived the rel-

evant common results from all studies and rated them

as more signiﬁcant toward other aspects we found.

Demands derived out of the interview series done by

(Denger et al., 2012) delivered the additional reduc-

tion criteria regarding the ﬁnal choice of the eval-

uation feature set. Table 1 shows the features set

www.google.com

http://www.sinequa.com

http://www.conweaver.de/

http://www.smartlogic.com/

http://www.intraﬁnd.com

Table 1: Extracted evaluation metrics.

Domain Feature

A. Indexing 1 Structured Data

(e.g databases, xml, MS Excell)

2 Semistructured and unstruc-

tured data

(e.g. e-mail, pdf, MS Word, MS

Powerpoint)

3 Persons

(e.g. contacts)

4 Media data

(videos, models, pictures etc.)

5 Meta data

6 Different sources

(ﬁle system, databases, social

media, web)

B. Rights 1 Early Binding

(integrative approach)

2 Sticky Policies

(item speciﬁc access)

C. Analytics 1 Semantic

2 Natural Language Processing

3 Statistic

4 Text Mining

D. Results 1 Classiﬁcation

(result type distinction)

2 Diversity

(diff. types at same time)

3 Locality

(location reproducible)

4 Results personalisation

5 Performance and scalability

(realtime, Big Data)

6 Alternative result visualisation

(diff. visualisation)

7 360 degree view

8 Navigation facets - search his-

tory

E. Search 1 Keywords

2 Query Expansion (QE)

3 Boolean operators

4 Wildcards

5 Case sensitivity

6 Different search modi

F. OS 1 Windows

2 Linux, Unix

3 Other

G. Support 1 Forum/Online Portal

2 E-mail

3 Chat

4 Hotline

5 Other

H. Languages 1 English

2 German

3 Other

I. Integration 1 Portals

2 Applications

3 CMS/DMS/ERP

4 Search engines

5 Other platforms

J. Administr. 1 Installation

2 Conﬁguration

3 Indexing control

APreliminaryShortSurveyofStateoftheArtEnterpriseSearchEnginesforFutureWorkPlace

627

that has been extracted as the ﬁnal version of litera-

ture study. As evaluation source we used white pa-

pers, ofﬁcial web pages, product sheets (IntraFind,

2013a; IntraFind, 2013b; ConWeaver, 2013; Dirsch-

Weigand and Schmidt, 2006; Sinequa, 2013; IBM,

2013; Smartlogic, 2013b; Smartlogic, 2013a) other

comparisons as well all accessible product facts that

can be found via Web. In the reference part of this

paper only ofﬁcial sources has been referenced. The

evaluation process will use value + for full support, o

for partial support, ˜ if uncertain information is only

available,- for none support and ? if no information

could be found. For legend refer Table 2.

Table 2: Evaluation symbols legend.

Symbol Meanining

+ full support

o partial support

˜ uncertain information

- none support

? no information found

4 RESULTS AND EVALUATION

The idea was to evaluate technical, functional, in-

tegral and user oriented features interesting for a

Knowledge Worker at Future Workplace. Results

on evaluation has been summarized in Table 3 and

compared in Figure 1. Considering A.Indexing,

C.Analytics, E:Search and D.Results feature overlap-

ping between the enterprise search engines is quite

high. Methodologies of content analysis differ, how-

ever the coverage of feature remains the same. For in-

stance Sinequa and ConWeaver reside more strongly

semantic knowledge networks while Semaphore and

IntraFind rely on Concept and Topic Maps. The 360

degree view remains typical for Sinequa as well as the

integrated Early Binding access rights support. All of

the search engines reveal in evaluation to be highly

integrative and modular (refer domain I.Integration).

Figure 1: Preliminary feature set comparision.

Some of them more in a generic way, the other ones in

customized sense, as in case of IntraFind where tailor

made solutions are considered as integration. Multi-

ple language support is claimed to be an extra good

integrated feature by all of the providers. Semaphore

hand in hand with Sinequa seems to have the most

professional G.Support, while Sinequa as only prod-

uct offers an indexing J.Administration. Although

Performance and Scalability (D.5) is characterized as

integral part of the product by each of the providers,

most impressive information has been delivered from

Sinequa

. Current installations deal with hundreds

of millions of documents. Highest divergence in re-

sults occurs in domains F.OS, J.Administration and

B.Rights.

5 CONCLUSIONS

Despite the awareness that a literature study can un-

veil only partly aspects of certain technical systems,

results of this survey can be still interpreted in differ-

ent ways. Choosing the best solution in given context

is no valid inference. This work is more overview

then rating. The main beneﬁt of this contribution re-

lies on extracted feature set which can be re-used and

reﬁned for further evaluations regarding the search

requirements for Future Workplace concepts where

search as integral part plays a very decisive role. In

the future work we are aiming at reﬁnement of fea-

ture set and expanding the survey on other enterprise

search providers. As stated in the introductory part all

evaluations rely on the information provided by the

vendors via web pages, product sheets and white pa-

pers as well from the one very massively elaborated

literature comparison done by (Gronau, 2005). The

authors made no technical evaluation of described so-

lutions which means that this evaluation is not in-

tended to be exhaustive and claims not to be complete.

ACKNOWLEDGEMENTS

The authors would like to acknowledge the ﬁnancial

support of the ”COMET K2 - Competence Centres for

Excellent Technologies Programme” of the Austrian

Federal Ministry for Transport, Innovation and Tech-

nology (BMVIT), the Austrian Federal Ministry of

Economy, Family and Youth (BMWFJ), the Austrian

Research Promotion Agency (FFG), the Province of

http://www.sinequa.com/en/page/solutions/

big-data.aspx

WEBIST2013-9thInternationalConferenceonWebInformationSystemsandTechnologies

628

Table 3: Survey results.

Sinequa ConWeaver Semaphore IntraFind

A.1 + + + +

A.2 + + + +

A.3 + + + +

A.4 + ? + ?

A.5 + + + +

A.6 + + + +

B.1 + - - -

B.2 - - ? -

C.1 + + + +

C.2 + + + +

C.3 + + o +

C.4 + + + +

D.1 + + + +

D.2 + + + +

D.3 + + + +

D.4 + + + +

D.5 + ˜ + +

D.6 + + + +

D.7 + ˜ ˜ ?

D.8 + + + o

E.1 + + + +

E.2 + + + +

E.3 + + + +

E.4 + + + +

E.5 + + + +

E.6 + + + +

F.1 + + + +

F.2 - o - +

F.3 - o - -

G.1 - - + -

G.2 + + + +

G.3 - - - -

G.4 + + + +

G.5 + + + +

H.1 + + + +

H.2 + + + +

H.3 + + + +

I.1 + + + o

I.2 + + + o

I.3 + + + o

I.4 o o + +

I.5 + ˜ + o

J.1 + ? + +

J.2 ? ? + +

J.3 ? ? ? +

Styria and the Styrian Business Promotion Agency

(SFG) .

REFERENCES

ConWeaver (2013). Conweaver white paper.

Website. http://www.conweaver.de/PDF/

ConWeaver Whitepaper.pdf.

Denger, A., Schmeja, M., and Stocker, A. (2012). Fu-

ture Workplace: eine Untersuchung sozio-technischer

Einﬂ

usse auf den Arbeitsplatz der Zukunft. Berichte

aus der Wirtschaftsinformatik. Shaker.

Dirsch-Weigand, A. and Schmidt, I. (2006). Conweaver:.

automatisierte wissensnetze fr die semantische suche.

Information - Wissenschaft & Praxis, 57(6/7):367–

371.

Drucker, P. (1959). The Landmarks of Tomorrow. Heine-

mann.

Gronau, N., editor (2005). Anwendungen und Systeme

ur das Wissensmanagement. Reihe Wirtschaftsinfor-

matik: technische und organisatorische Gestaltung-

soptionen. Gito-Verl., Berlin, 2. auﬂ. edition.

Hager, S. (1997). Evaluierung von Suchmaschinen fr den

Einsatz im BMWIntranet. Technische Universit

unchen.

IBM (2013). Ibm partnerworld, global solutions di-

rectory, sinequa description. Website. http://

www-304.ibm.com/partnerworld/gsd/solutiondetails.

do?solution=32061&expand=true&lc=en.

IntraFind (2013a). iﬁnder enterprise search un-

ternehmensweit informationen ﬁnden. http://

www.intraﬁnd.de/de/produkte/iﬁnder.

IntraFind (2013b). iﬁnder enterprise search, prod-

uct data sheet. Website. http://www.intraﬁnd.de/

tl ﬁles/documents/Produkte/INTRAFIND Datenblatt

iFinder4 7 DE 022013.pdf.

Mukherjee, R. and Mao, J. (2004). Enterprise search:

Tough stuff. Queue, 2(2):36–46.

Sinequa (2013). Product overview. Website. http://

www.sinequa.com/en/page/product/product.aspx.

Smartlogic (2013a). Semantic enhancement server white

paper. Website. http://www.smartlogic.com/request-

download? f=35/SL SES WPNT0012011 EN.pdf.

Smartlogic (2013b). Semaphore overview a smartlogic

white paper. Website. http://www.smartlogic.com/

requestdownload?f=121/SL SemaphoreOverview

WPNT0052011 EN.PDF.

Stocker, A., Denger, A., H

ubler, A., Ruckriegel, H., Maletz,

M., and Klimisch, M. (2012). Arbeitsplatz der zukunft

mit fallstudien von bmw und avl. HMD - Praxis

Wirtschaftsinform., 287.

Ulbrich, A., Kraker, P., and Luidolt, C. (2009). Ergebnisse

einer untersuchung von anbietern von suchmaschinen-

technologien. In Hinkelmann, K. and Wache, H., ed-

itors, Wissensmanagement, volume 145 of LNI, pages

118–127. GI.

APreliminaryShortSurveyofStateoftheArtEnterpriseSearchEnginesforFutureWorkPlace

629