Cover Page


B.3.2.b.1 WT1: Work package list



Yüklə 0,67 Mb.
səhifə10/12
tarix31.12.2018
ölçüsü0,67 Mb.
#88621
1   ...   4   5   6   7   8   9   10   11   12

B.3.2.b.1 WT1: Work package list




WPiNo.

Work Package

Title


Lead

Participant No. (ii)



Lead Participant

Short name (iii)



Total person months

per WP (iv)



Start

Month (v)



End

Month (vi)



1

Management

1

XRCE

30

0

36

2

Language Resources

1

XRCE

30

0

14

3

Log Analysis

5

OD

44

3

10

4

Algorithms tuning

4

UNITN

55

6

36

5

Machine Translation for Queries

3

UVA

59,5

0

36

6

Optimisation

5

OD

74,14

0

36

7

Evaluation

8

UBER

58,75

12

36

8

Business sustainability

2

CELI

47,1

0

36

9

Dissemination

1

XRCE

37

0

36




TOTAL







435,49






Given the nature of the project, it was decided to further articulate the WPs into tasks, in order to capture fine grained dependencies. Also allocations were decided “by task” and WP allocations are just the sum of all relative task allocation. The following matrix represents Task/Partner allocation (Tasks identifiers are explained in the relative work package):







B.3.2.b.2 WT2: Deliverables list


Grey marked rows are deliverable required by Annex X, Yellow marked rows are pure “project” deliverable, unmarked rows are deliverables produced as regular technical or commercial/professional publications


Deliverable
No

Deliverable name

WP

No.



Nature

Dissemination
level

Delivery
date (proj. month)

D1.1

Consortium Agreement

1

R

CO

3

D9.1.

Dissemination and awareness plan


9

R

CO

3

D9.2

Project web site and power point presentation

9

P

PU

3

D1.2.a

Periodic Activity Report

1

R

CO

6

D1.3.a

Progress Report

1

R

CO

6

D6.1

System design

6

R

PU

3

D2.1

Wrapped services of NLP

2

P

PU

6

D2.3

Language identification component

2

P

PU

6

D8.1

Business Plan

8

R

CO

6

D3.1

Component for harvesting, normalizing and storing logs

3

P

PU

9

D3.2

Integration with main-stream log analysis systems

3

P

PU

9

D3.3

Data Mining for GALATEAS-enriched logs

3

P

PU

9

D4.1

Definition of the interfaces for the al the algorithms

4

P

PU

9

D1.2.b

Periodic Activity Report

1

R

CO

12

D1.3.b

Progress Report

1

R

CO

12

D1.4.a

Annual Report

1




PU

12

D4.2

Topic Computation

4

P

PU

12

D4.3:

Clustering

4

P

PU

12

D4.4

Language-based Click through


4

P

PU

12

D5.1.

MOSES parametrization approach (m.12)

5

P

PU

12

D6.2

First Version of the integrated architecture+ LangLog WS

6

D

PU

12

D4.5

Extended TLike

4

P

PU

15

D7.1

First evaluation report of Topic Computation and TLIKE

7

R

PU

15

D1.2.c

Periodic Activity Report

1

R

CO

18

D1.3.c

Progress Report

1

R

CO

18

D2.2

NE extraction service

2

P

PU

18

D5.2

Integration of MT with LangLog flow.

5

P

PU

18

D8.2

User study on search episodes

8

R

PU

18

D9.3

GALATEAS services web site

9

D

PU

18

D9.4

First report on dissemination activities

9

R

CO

18

D1.4.b

Annual Report

1

R

PU

24

D1.3.d

Progress Report

1

R

CO

24

D1.2.d

Periodic Activity Report

1

R

CO

24

D6.3

Final Version of the integrated architecture + QueryTrans WS.

6

D

PU

24


D6.4

Configuration GUI

6

P

PU

24

D7.3

First Evaluation of Query Translation

7

R

PU

24

D9.5

Project showcase

9

P

PU

24

D1.2.e

Periodic Activity Report

1

R

CO

30

D1.3.e

Progress Report

1

R

CO

30

D8.3

Intermediate report on customer acquisition activities

8

R

CO

30

D1.2.f

Periodic Activity Report

1

R

CO

36

D1.3.f

Progress Report

1

R

CO

36

D1.5

Final Report

1

R

PU

36

D4.6

Final version of tuned algorithms

4

P

PU

36

D5.3.

MT infrastructure

5

P

PU

36

D7.2

Final evaluation report of Topic Computation and TLIKE

7

R

PU

36

D7.4

Final Evaluation of Query Translation

7

R

PU

36

D8.4

Final report on customer acquisition activities

8

R

CO

36

D9.6

Second report on dissemination activities

9

R

CO

36

In addition, the Consortium undertakes to perform the tasks, submit the deliverables and organise respectively take part in the events described in Appendix X to this Description of Work.
R = Report, P = Prototype, D = Demonstrator, O = Other PU = Public

PP = Restricted to other programme participants (including the Commission Services)

RE = Restricted to a group specified by the consortium (including the Commission Services)

CO = Confidential, only for members of the consortium (including the Commission Services)

There is a strict relationship between deliverables and tasks: in the following table. Emphasis is put on this connection (which is sometimes hidden by different numbering) and specifies the main responsibilities of partners:




T2.1

D2.1

NLP Service Wrapping

Xerox will coordinate the task. Xerox and CELI will integrate their NLP module in the proposed architecture.

T.2.2

D2.2

NE extraction for queries

The main responsible will be UNITN, which will adapt language-independent methods. Xerox will support UNITN by providing NLP capabilities, when required.

T2.3

D2.3

Language Identification Component

UNITN will be the main responsible for this task. CELI will provide configuration for three of the selected languages

T3.1

D3.1

Log Management

OD will be the responsible for this task. GONET will implement log harvesting. OD will implement DB design and population, as well as log normalization.

T3.2

D3.2

Log Information disclosure

OD will be the responsible for this task. GONET will implement integration with AWSTATS and WEBTRENDS. OD will implement integration with Mondrian and BO.

T4.1

D4.1

Clustering

The responsible of the task is UNITN. UNITN will configure the clustering algorithms for queries. XEROX will provide language support.

T4.2

D4.2

Topic Computation

The responsible of the task is UNITN. UNITN will implement the click-through classification algorithm; OD will configure the language based algorithm

T4.3

D4.3

Extended TLike

CELI is responsible of the task and of the extension of TLIKE: UBER will provide data for testing and evaluation.

T4.4

D4.4

Algorithm Optimization

This task will be lead by UNITN: OD and CELI will contribute to optimize their own algorithm. UBER will provide data.

T5.1

D5.1

MT Adaptation

UVA will lead this task. XRCE will take care of configuring Moses for languages with dependency information, whereas UVA will take care of “baseline” cases. CELI will integrate MT with dictionary based approaches

T5.2

D5.2

Integration with LangLog and domain adaptation

UVA will lead this task. UVA and XRCE will implement methodologies to achieve better tailoring of the MT system across domains. CELI will realize the integration with LangLog

T5.3

D5.3

MT Infrastructure set up

CELI is responsible for this task. CELI will choose and parametrize the adequate infrastructure: UVA and XRCE will contribute to optimize their components accordingly.

T6.1

D6.1

System design

GONET will coordinate the task. All technical industrial partners are involved in order to design interfaces and standards for each component of the system.

T6.2

D6.2

First version of the integrated service

OD is responsible of this task. CELI will contribute in managing the interaction between local and remote services.

T6.3

D6.2

LangLog service

OD is responsible for this task and unique executor.

T6.4

D6.3

Final version of the integrated Architecture.

OD is the responsible for this task and the main executor. CELI will be responsible of the transformation of local computations into web services.

T6.5

D6.3

QueryTrans Service

XRCE is responsible for this task and unique executor.

T 6.6

D6.4

Configuration GUI

XRCE is responsible for this task: It will provide specification and guidelines to GONET which will take care of the GUI implementation.

T6.7




Service Maintenance and Adaptation

GONET is responsible for this task. CELI and XRCE will take care of adaptation to new domain. GONET will have to maintain a scalable infrastructure. OD will take care on parametrization activity for log disclosure

T7.1

D7.1/

D7.2


Evaluation of Topic Computation

UBER is responsible for this task. UNITN will be responsible for the evaluation infrastructure. BAL and UBER will provide manual annotations and will disclose data for the evaluation. OD will evaluate non functional aspects and involve a third party library.

T7.2

D7.1/

D7.2


Evaluation of TLike

UBER is responsible for this task. UNITN will be responsible for the evaluation infrastructure. BAL, UBER and UVA will provide manual annotations and will disclose data for the evaluation. OD will evaluate non functional aspects and involve a third party library.

T7.3

D7.3/

D7.4


Evaluation of Query Translation

BAL is responsible for this task and will set up an evaluation environment for their digital collection. Both UBER and BAL will set up gold standard for evaluation. OD will evaluate non functional aspects and involve a third party library.

T8.1

D8.1

Business plan creation

CELI is responsible for this task. It will ask the other participating partners focused contributions on relevant aspects.

T8.2

D8.2

User study on search episodes

UNITN is responsible on the task. It will ask contribution from UBER and BAL. CELI is in this task to provide technical assistance

T8.3

D8.3/

D8.4


Customer Acquisition

CELI is the responsible of this task and the main designer of opt in strategy. GONET, XRCE and OD, will implement these strategies in their respective fields of influence.

T9.1

D9.1/

D9.2/


D9.4/

D9.5/


D9.6

Scientific Community Involvement

XRCE is the responsible of this task. All the partners will contribute to the activities as planned in D9.1

T9.2

D9.1/

D9.3/


D9.4/

D9.5/


D9.6

Business Community Involvement

CELI is the responsible of this task. All the partners will contribute to the activities as planned in D9.1



Yüklə 0,67 Mb.

Dostları ilə paylaş:
1   ...   4   5   6   7   8   9   10   11   12




Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©muhaz.org 2024
rəhbərliyinə müraciət

gir | qeydiyyatdan keç
    Ana səhifə


yükləyin