text
stringlengths
265
1.11k
### Instruction: Extract the cause and effect entities from the following sentence: These systems employ sound signals instead of ultrasound signal to avoid the signal attenuation ### Assistant: NEGATION: instead of VARIABLE: These systems CONDITION: ultrasound signal CAUSE_1: These systems employ sound signals CAUSE_2: instead of ultrasound signal EFFECT_1: to avoid the signal attenuation ### END
### Instruction: Extract the cause and effect entities from the following sentence: Because the devices use a radio (broadcast) communications system, they do not have to be in the LoS of each other and the synchronization is avoided ### Assistant: KEYWORD: Because VARIABLE: the synchronization CONDITION: have to be in the LoS of each other NOT_RELEVANT: (broadcast) CAUSE_1: communications system NEGATION: is avoided AND: and EFFECT_2: the synchronization EFFECT_1: they do not have to be in the LoS of each other ### END
### Instruction: Extract the cause and effect entities from the following sentence: By reading and identifying the tags (Assuming that the readers are placed in a well known position) the distance or the placement of nodes estimation is calculated ### Assistant: CONDITION: is calculated AND: and VARIABLE: the placement of nodes estimation NOT_RELEVANT: (Assuming that the readers are placed in a well known position) OR: or CAUSE_1: reading CAUSE_2: identifying the tags EFFECT_1: the distance EFFECT_2: the placement of nodes estimation is calculated KEYWORD: By ### END
### Instruction: Extract the cause and effect entities from the following sentence: To compute the distances to anchors, the trilateration make use of the previously reviewed distance estimation techniques ### Assistant: CONDITION: make use of the previously reviewed distance estimation techniques VARIABLE: the trilateration EFFECT_1: To compute the distances to anchors CAUSE_1: the trilateration make use of the previously reviewed distance estimation techniques ### END
### Instruction: Extract the cause and effect entities from the following sentence: When a node desires to compute its position but the anchors are not in the node’s range, it makes use of its neighbor nodes as relays to reach the beacon nodes and eventually compute its current position ### Assistant: KEYWORD: When VARIABLE: its current position CONDITION: compute AND: and CAUSE_1: a node desires to compute its position NEGATION: not CAUSE_2: the anchors are not in the node’s range NOT_RELEVANT: eventually EFFECT_1: it makes use of its neighbor nodes as relays to reach the beacon nodes EFFECT_2: compute its current position ### END
### Instruction: Extract the cause and effect entities from the following sentence: Generally, to compute the position two angles plus a distance between two well known anchors are needed (2 Dimensions - for 3 Dimensions also the azimuth is used) ### Assistant: NOT_RELEVANT: (2 Dimensions - for 3 Dimensions also the azimuth is used) VARIABLE: a distance between two well known anchors CONDITION: are needed EFFECT_1: to compute the position AND: plus CAUSE_1: two angles CAUSE_2: a distance between two well known anchors are needed ### END
### Instruction: Extract the cause and effect entities from the following sentence: In this case context tags are used to implement the context-aware technologies ### Assistant: CONDITION: are used VARIABLE: context tags EFFECT_1: to implement the context-aware technologies KEYWORD: In this case CAUSE_1: context tags are used ### END
### Instruction: Extract the cause and effect entities from the following sentence: Also, middleware software architectures should offer mechanisms to achieve adaptive and efficient system resources use, in order to prolong the sensor network's life ### Assistant: CONDITION: to prolong VARIABLE: the sensor network's life NOT_RELEVANT: Also CAUSE_1: middleware software architectures should offer mechanisms AND: and EFFECT_1: to achieve adaptive EFFECT_2: efficient system resources use KEYWORD: in order EFFECT_3: to prolong the sensor network's life. ### END
### Instruction: Extract the cause and effect entities from the following sentence: A global WSN’s behavior is programmed at a high-level specification that enables node concerned about dealing with low-level ### Assistant: KEYWORD: that CONDITION: is programmed at a high-level specification EFFECT_1: enables node concerned about dealing with low-level VARIABLE: node concerned about dealing CAUSE_1: A global WSN’s behavior is programmed at a high-level specification ### END
### Instruction: Extract the cause and effect entities from the following sentence: In this sense, the middleware layer in Wireless Sensor Networks has to implement some mechanisms in order to reach efficient deployments of context applications over Wireless Sensor Networks ### Assistant: KEYWORD: in order CONDITION: has to implement some mechanisms VARIABLE: the middleware layer in Wireless Sensor Networks EFFECT_1: to reach efficient deployments of context applications over Wireless Sensor Networks NOT_RELEVANT: In this sense CAUSE_1: the middleware layer in Wireless Sensor Networks has to implement some mechanisms ### END
### Instruction: Extract the cause and effect entities from the following sentence: The contextual model implemented by this proposal allows using different reasoning mechanisms like first order logic and temporal logic ### Assistant: KEYWORD: allows VARIABLE: different reasoning mechanisms CONDITION: using CAUSE_1: The contextual model implemented by this proposal EFFECT_1: using different reasoning mechanisms NOT_RELEVANT: like first order logic and temporal logic ### END
### Instruction: Extract the cause and effect entities from the following sentence: The context interpretation is achieved by using an interpreter based on fuzzy rules called CIBFR (Context Interpreter Based on Fuzzy Rule) ### Assistant: KEYWORD: by VARIABLE: an interpreter CONDITION: using EFFECT_1: The context interpretation is achieved CAUSE_1: using an interpreter based on fuzzy rules NOT_RELEVANT: called CIBFR (Context Interpreter Based on Fuzzy Rule). ### END
### Instruction: Extract the cause and effect entities from the following sentence: Otherwise, weak WSN devices enable higher deployment densities and increase network lifetime as they are cheaper and consume less power ### Assistant: CONDITION: consume less power VARIABLE: they CAUSE_1: weak WSN devices EFFECT_1: enable higher deployment densities AND: and EFFECT_2: increase network lifetime NOT_RELEVANT: Otherwise KEYWORD: as CAUSE_2: they are cheaper CAUSE_3: consume less power ### END
### Instruction: Extract the cause and effect entities from the following sentence: By integrating devices with different resources and capabilities, a heterogeneous WSN can combine the advantages of both powerful and weak devices ### Assistant: CONDITION: powerful VARIABLE: a heterogeneous WSN AND: of both CAUSE_1: integrating devices with different resources EFFECT_1: powerful EFFECT_2: weak devices KEYWORD: By CAUSE_2: capabilities ### END
### Instruction: Extract the cause and effect entities from the following sentence: New programming models are necessary to simplify WSN application development and increase overall network utility ### Assistant: VARIABLE: overall network utility CONDITION: increase CAUSE_1: New programming models are necessary EFFECT_1: to simplify WSN application development AND: and EFFECT_2: increase overall network utility ### END
### Instruction: Extract the cause and effect entities from the following sentence: These services are dynamically discovered and used at run-time, enabling applications to be platform-independent and adapt to network dynamics ### Assistant: CONDITION: used at run-time VARIABLE: These services EFFECT_1: enabling applications to be platform-independent AND: and EFFECT_2: adapt to network dynamics CAUSE_1: These services are dynamically discovered CAUSE_2: used at run-time ### END
### Instruction: Extract the cause and effect entities from the following sentence: While service-oriented computing is widely used on the Internet, adopting it to WSNs is non-trivial due to the extremely limited resources available and highly dynamic nature ### Assistant: KEYWORD: due to VARIABLE: the CONDITION: highly dynamic CAUSE_1: service-oriented computing is widely used on the Internet EFFECT_1: adopting it to WSNs is non-trivial AND: and CAUSE_2: the extremely limited resources available CAUSE_3: highly dynamic nature ### END
### Instruction: Extract the cause and effect entities from the following sentence: Services are able to maintain state, provide multiple methods, and have their own thread of control, enabling them to operate in parallel with task ### Assistant: CONDITION: provide multiple methods VARIABLE: Services EFFECT_1: enabling them to operate in parallel with task. AND: and CAUSE_1: Services are able to maintain state CAUSE_2: provide multiple methods ### END
### Instruction: Extract the cause and effect entities from the following sentence: To achieve the previous exposed goals, SNMS develops a gathering tree to collect network health information, introducing a minimal impact on memory and network traffic ### Assistant: CONDITION: introducing VARIABLE: a minimal impact on memory EFFECT_1: To achieve the previous exposed goals AND: and CAUSE_1: SNMS develops a gathering tree to collect network health information CAUSE_3: network traffic CAUSE_2: introducing a minimal impact on memory ### END
### Instruction: Extract the cause and effect entities from the following sentence: On the other hand, SNMS network management function is limited to passive monitoring only, requiring human managers to submit queries and perform post-mortem analysis of management data ### Assistant: NOT_RELEVANT: On the other hand VARIABLE: human managers CONDITION: perform post-mortem analysis of management data AND: and CAUSE_1: SNMS network management function is limited to passive monitoring only EFFECT_1: requiring human managers to submit queries EFFECT_2: perform post-mortem analysis of management data ### END
### Instruction: Extract the cause and effect entities from the following sentence: To interconnect different SSNs is proposed the use of gateways ### Assistant: CONDITION: the use of VARIABLE: gateways EFFECT_1: To interconnect different SSNs KEYWORD: is proposed CAUSE_1: the use of gateways ### END
### Instruction: Extract the cause and effect entities from the following sentence: Event-centric wireless sensor networks (EC-WSNs) differ from common WSNs, where the communications are triggered either by an on-demand or a sink-based process ### Assistant: KEYWORD: where VARIABLE: Event-centric wireless sensor networks CONDITION: differ from common WSNs OR: or CAUSE_1: the communications are triggered either by an on-demand CAUSE_2: a sink-based process NOT_RELEVANT: (EC-WSNs) EFFECT_1: Event-centric wireless sensor networks (EC-WSNs) differ from common WSNs ### END
### Instruction: Extract the cause and effect entities from the following sentence: Indeed, in this case, the communication is triggered only when an event is detected in the immediate sensing-range of a node ### Assistant: NOT_RELEVANT: Indeed, in this case KEYWORD: only when VARIABLE: the communication CONDITION: is triggered CAUSE_1: an event is detected in the immediate sensing-range of a node EFFECT_1: the communication is triggered ### END
### Instruction: Extract the cause and effect entities from the following sentence: Attacks from malicious motes that are not part of the network and try to join it without authorization, are often done applying cryptanalysis techniques and attacking physically the device ### Assistant: VARIABLE: the device CONDITION: applying NEGATION: without AND: and NOT_RELEVANT: are often done CAUSE_2: applying cryptanalysis techniques CAUSE_3: attacking physically the device KEYWORD: that CAUSE_1: Attacks from malicious motes EFFECT_1: are not part of the network EFFECT_2: try to join it EFFECT_3: without authorization ### END
### Instruction: Extract the cause and effect entities from the following sentence: Motes are part of the network and they may attack if they have been compromised by an attacker that has manipulated them [66] ### Assistant: KEYWORD: that AND: and VARIABLE: they CONDITION: has manipulated them EFFECT_1: they may attack CAUSE_1: they have been compromised by an attacker NOT_RELEVANT: [66] CAUSE_2: has manipulated them ### END
### Instruction: Extract the cause and effect entities from the following sentence: Thus also studies made by our researchers [78], has been shown that the use of the AES algorithm reduces half the batteries life ### Assistant: KEYWORD: has been shown that NOT_RELEVANT: Thus also studies made by our researchers [78], CONDITION: reduces half VARIABLE: the batteries life CAUSE_1: the use of the AES algorithm EFFECT_1: reduces half the batteries life ### END
### Instruction: Extract the cause and effect entities from the following sentence: The traditional DES (Data Encryption Standard) uses many computational resources and therefore it isn’t recommended [79] for use in WSN ### Assistant: KEYWORD: therefore VARIABLE: it NOT_RELEVANT: [79] CONDITION: for use in WSN NEGATION: isn’t CAUSE_1: The traditional DES (Data Encryption Standard) uses many computational resources EFFECT_1: for use in WSN AND: and ### END
### Instruction: Extract the cause and effect entities from the following sentence: In digital mobile networks, location data giving the geographic position of the terminal equipment of the mobile user are processed to enable the transmission of communications ### Assistant: CONDITION: are processed VARIABLE: digital mobile networks EFFECT_1: to enable the transmission of communications CAUSE_2: location data giving the geographic position of the terminal equipment of the mobile user are processed KEYWORD: In CAUSE_1: digital mobile networks ### END
### Instruction: Extract the cause and effect entities from the following sentence: Sender and receiver share a secret key to generate and verify a MAC (Message Authentication Code) [72] ### Assistant: CONDITION: share a secret key AND: and VARIABLE: receiver NOT_RELEVANT: (Message Authentication Code) [72] EFFECT_1: to generate EFFECT_2: verify a MAC CAUSE_1: Sender CAUSE_2: receiver share a secret key ### END
### Instruction: Extract the cause and effect entities from the following sentence: However it may cause a bottleneck process from the sensor nodes if there is excessive traffic on the network ### Assistant: KEYWORD: if there CONDITION: may cause a bottleneck process from the sensor nodes CAUSE_1: is excessive traffic on the network VARIABLE: traffic on the network EFFECT_1: it may cause a bottleneck process from the sensor nodes NOT_RELEVANT: However ### END
### Instruction: Extract the cause and effect entities from the following sentence: This mechanism based on multiple MAC messages and requires that network nodes have a key ring ### Assistant: KEYWORD: and requires that VARIABLE: This mechanism CONDITION: based on multiple MAC messages EFFECT_1: network nodes have a key ring CAUSE_1: This mechanism based on multiple MAC messages ### END
### Instruction: Extract the cause and effect entities from the following sentence: To authenticate a message, the source node generate a list of MAC based on their keys, and added them to the message ### Assistant: CONDITION: added them to the message VARIABLE: the source node EFFECT_1: To authenticate a message AND: and CAUSE_1: the source node generate a list of MAC based on their keys CAUSE_2: added them to the message ### END
### Instruction: Extract the cause and effect entities from the following sentence: The drawback is that in order to operate the device, the computer needs to be running ### Assistant: KEYWORD: in order CONDITION: needs to be running VARIABLE: the computer NOT_RELEVANT: The drawback is that EFFECT_1: to operate the device CAUSE_1: the computer needs to be running ### END
### Instruction: Extract the cause and effect entities from the following sentence: Networked device: Arduino boards can participate in a network, either using Ethernet or ZigBee shields with the corresponding libraries ### Assistant: OR: or VARIABLE: Arduino boards CONDITION: can participate in a network CAUSE_1: using Ethernet CAUSE_2: ZigBee shields with the corresponding libraries EFFECT_1: Arduino boards can participate in a network NOT_RELEVANT: Networked device ### END
### Instruction: Extract the cause and effect entities from the following sentence: A huge number of embedded devices today run *NIX-like operating systems such as GNU/Linux and are open or hacked so that they can run custom software ### Assistant: KEYWORD: so that VARIABLE: A huge number of embedded devices CONDITION: hacked EFFECT_1: they can run custom software AND: and OR: or NOT_RELEVANT: such as GNU/Linux CAUSE_1: A huge number of embedded devices today run *NIX-like operating systems CAUSE_2: are open CAUSE_3: hacked ### END
### Instruction: Extract the cause and effect entities from the following sentence: Eventing: if devices are meant to receive commands and notify events, a callback mechanism should be implemented ### Assistant: KEYWORD: if VARIABLE: a callback mechanism CONDITION: should be implemented AND: and NOT_RELEVANT: Eventing CAUSE_1: devices are meant to receive commands CAUSE_2: notify events EFFECT_1: a callback mechanism should be implemented ### END
### Instruction: Extract the cause and effect entities from the following sentence: As an example, an increasing number of hobbyists that want wireless battery-powered devices buy ready-to-use ZigBee modules [14] ### Assistant: VARIABLE: an increasing number of hobbyists CONDITION: buy ready-to-use ZigBee modules CAUSE_1: want wireless battery-powered devices NOT_RELEVANT: As an example EFFECT_1: buy ready-to-use ZigBee modules KEYWORD: that ### END
### Instruction: Extract the cause and effect entities from the following sentence: These modules are easy to use as they behave like simple modems (communication uses serial and configuration is done through Hayes AT commands) ### Assistant: KEYWORD: as VARIABLE: These modules CONDITION: are easy to use NOT_RELEVANT: (communication uses serial and configuration is done through Hayes AT commands) EFFECT_1: These modules are easy to use CAUSE_1: they behave like simple modems ### END
### Instruction: Extract the cause and effect entities from the following sentence: A radio is considered in “Shunting Mode” when it is prepared to receive shunting emergency calls but cannot receive train emergency calls ### Assistant: KEYWORD: when VARIABLE: A radio CONDITION: is considered in “Shunting Mode” AND: but NEGATION: cannot CAUSE_1: it is prepared to receive shunting emergency calls CAUSE_2: cannot receive train emergency calls EFFECT_1: A radio is considered in “Shunting Mode” ### END
### Instruction: Extract the cause and effect entities from the following sentence: A radio is considered in “Train Mode” when it is prepared to receive train emergency calls but cannot receive shunting emergency calls ### Assistant: KEYWORD: when VARIABLE: A radio CONDITION: is considered in “Train Mode” AND: but NEGATION: cannot CAUSE_1: it is prepared to receive train emergency calls CAUSE_2: cannot receive shunting emergency calls EFFECT_1: A radio is considered in “Train Mode” ### END
### Instruction: Extract the cause and effect entities from the following sentence: The EIRENE Functional Requirements Specification defines a set of requirements which a railway radio system should comply with, in order to facilitate international interoperability between national railways and manufacturing economies of scale ### Assistant: KEYWORD: which CONDITION: should comply with VARIABLE: a railway radio system EFFECT_1: to facilitate international interoperability between national railways and manufacturing economies of scale CAUSE_2: a railway radio system should comply with CAUSE_1: The EIRENE Functional Requirements Specification defines a set of requirements ### END
### Instruction: Extract the cause and effect entities from the following sentence: Any user who is not within the list of allowed EIRENE users shall not be able to gain access to any of the functions and services provided by the network ### Assistant: VARIABLE: Any user NEGATION: shall not CONDITION: services provided by the network CAUSE_1: is not within the list of allowed EIRENE users EFFECT_1: shall not be able to gain access to any of the functions AND: and EFFECT_2: services provided by the network KEYWORD: who ### END
### Instruction: Extract the cause and effect entities from the following sentence: The level of coverage should be at least 95% of the time over 95% of the designated coverage area for a radio installed in a vehicle with an external antenna ### Assistant: KEYWORD: for VARIABLE: The level of coverage CONDITION: should be at least 95% of the time over 95% of the designated coverage area AND: with CAUSE_1: a radio installed in a vehicle CAUSE_2: an external antenna EFFECT_1: The level of coverage should be at least 95% of the time over 95% of the designated coverage area ### END
### Instruction: Extract the cause and effect entities from the following sentence: Any environmental and physical requirements stated may be superseded by national requirements if the national standards provide a higher level of environmental and physical protection ### Assistant: KEYWORD: if VARIABLE: physical requirements stated CONDITION: may be superseded by national requirements AND: and CAUSE_1: the national standards provide a higher level of environmental CAUSE_2: physical protection EFFECT_1: Any environmental EFFECT_2: physical requirements stated may be superseded by national requirements ### END
### Instruction: Extract the cause and effect entities from the following sentence: If the system is not able to connect the call, an audible and visual indication shall be provided to the driver that the call was not received by the controller ### Assistant: VARIABLE: visual indication NEGATION: is not CONDITION: shall be provided to the driver that the call was not received by the controller KEYWORD: If CAUSE_1: the system is not able to connect the call AND: and EFFECT_1: an audible EFFECT_2: visual indication shall be provided to the driver that the call was not received by the controller ### END
### Instruction: Extract the cause and effect entities from the following sentence: Additional information may also be provided if available from external systems ### Assistant: KEYWORD: if VARIABLE: Additional information CONDITION: available from external systems CAUSE_1: available from external systems EFFECT_1: Additional information may also be provided ### END
### Instruction: Extract the cause and effect entities from the following sentence: It shall not be possible for any General purpose radio user to send a Railway emergency call by dialling a short code or telephone number ### Assistant: VARIABLE: a short code OR: or CAUSE_1: any General purpose radio user CAUSE_2: telephone number NEGATION: not be possible NOT_RELEVANT: It shall KEYWORD: by CONDITION: to send a Railway emergency call EFFECT_1: to send a Railway emergency call ### END
### Instruction: Extract the cause and effect entities from the following sentence: The user may then talk whilst the PTT function is used ### Assistant: KEYWORD: then VARIABLE: The user CONDITION: talk CAUSE_1: the PTT function is used EFFECT_1: talk ### END
### Instruction: Extract the cause and effect entities from the following sentence: Changing the battery should not result in losing data stored in the radio ### Assistant: VARIABLE: data stored in the radio CONDITION: result in losing NEGATION: should not CAUSE_1: Changing the battery EFFECT_1: should not result in losing data stored in the radio ### END
### Instruction: Extract the cause and effect entities from the following sentence: If the ability to initiate Railway emergency calls is implemented in the radio, accidental initiation of a Railway emergency call shall be prevented ### Assistant: KEYWORD: If CONDITION: shall be VARIABLE: accidental initiation of a Railway emergency call CAUSE_1: the ability to initiate Railway emergency calls is implemented in the radio NEGATION: prevented EFFECT_1: accidental initiation of a Railway emergency call shall be prevented ### END
### Instruction: Extract the cause and effect entities from the following sentence: It shall be possible to change the battery without losing data stored in the radio ### Assistant: NEGATION: without CONDITION: change VARIABLE: the battery EFFECT_1: without losing data stored in the radio CAUSE_1: change the battery NOT_RELEVANT: It shall be possible to ### END
### Instruction: Extract the cause and effect entities from the following sentence: For activation, a maximum of 2 MMI actions should be allowed ### Assistant: KEYWORD: For VARIABLE: a maximum of 2 MMI actions CONDITION: should be allowed CAUSE_1: activation EFFECT_1: a maximum of 2 MMI actions should be allowed ### END
### Instruction: Extract the cause and effect entities from the following sentence: International standardisation of numbering plans is needed to ensure interworking between networks ### Assistant: VARIABLE: interworking between networks CONDITION: to ensure CAUSE_1: to ensure interworking between networks EFFECT_1: International standardisation of numbering plans is needed ### END
### Instruction: Extract the cause and effect entities from the following sentence: For certain functions, standardised telephone numbers shall be implemented ### Assistant: KEYWORD: For VARIABLE: standardised telephone numbers CONDITION: shall be implemented CAUSE_1: certain functions EFFECT_1: standardised telephone numbers shall be implemented ### END
### Instruction: Extract the cause and effect entities from the following sentence: To provide interoperability between the fixed railway networks within the EIRENE network, standardisation of UIC group numbering will be required ### Assistant: CONDITION: will be required VARIABLE: the EIRENE network EFFECT_1: To provide interoperability between the fixed railway networks CAUSE_2: standardisation of UIC group numbering will be required KEYWORD: within CAUSE_1: the EIRENE network ### END
### Instruction: Extract the cause and effect entities from the following sentence: During this recovery period, the system shall not permit the use of unverified functional numbers ### Assistant: KEYWORD: During VARIABLE: the system NEGATION: shall not CONDITION: permit the use of unverified functional numbers CAUSE_1: this recovery period EFFECT_1: the system shall not permit the use of unverified functional numbers ### END
### Instruction: Extract the cause and effect entities from the following sentence: If the mobile is in shunting mode, the emergency call button shall initiate a shunting emergency call, otherwise the call shall be a Train emergency call ### Assistant: KEYWORD: If VARIABLE: the call CONDITION: shall be a Train emergency call CAUSE_1: the mobile is in shunting mode EFFECT_1: the emergency call button shall initiate a shunting emergency call NEGATION: otherwise EFFECT_2: the call shall be a Train emergency call ### END
### Instruction: Extract the cause and effect entities from the following sentence: For this reason, we decided keeping them in a dedicate server to improve the platform performance ### Assistant: NOT_RELEVANT: we decided CONDITION: in a dedicate server VARIABLE: them EFFECT_1: to improve the platform performance CAUSE_1: keeping them in a dedicate server ### END
### Instruction: Extract the cause and effect entities from the following sentence: We also would like to develop a mobile application for the more common mobile systems (Android and iOS), in order to get a better response from the user ### Assistant: NOT_RELEVANT: We also would like KEYWORD: in order CONDITION: to develop VARIABLE: a mobile application for the more common mobile systems EFFECT_1: to get a better response from the user CAUSE_1: to develop a mobile application for the more common mobile systems ### END
### Instruction: Extract the cause and effect entities from the following sentence: One of the most important objectives is to keep all content in the platform hidden for public users, because we are working, in some cases, with underage people ### Assistant: KEYWORD: because VARIABLE: all content in the platform CONDITION: with underage people CAUSE_1: with underage people NOT_RELEVANT: in some cases EFFECT_1: to keep all content in the platform hidden for public users ### END
### Instruction: Extract the cause and effect entities from the following sentence: For improving the usability, we have planned to design custom styles for each age group, because of teenagers do not want to use a childish platform ### Assistant: KEYWORD: because of CONDITION: want to use a childish platform VARIABLE: teenagers EFFECT_1: improving the usability CAUSE_1: to design custom styles CAUSE_2: age group NOT_RELEVANT: we have planned NEGATION: do not CAUSE_3: teenagers do not want to use a childish platform ### END
### Instruction: Extract the cause and effect entities from the following sentence: We have also kept in mind that if a user likes our platform, she would like to share it with her friends, for this reason, we will develop an easy tool for social sharing ### Assistant: KEYWORD: for this reason NOT_RELEVANT: We have also kept in mind that VARIABLE: we CONDITION: will develop an easy tool for social sharing CAUSE_1: a user likes our platform EFFECT_1: she would like to share it with her friends EFFECT_2: we will develop an easy tool for social sharing ### END
### Instruction: Extract the cause and effect entities from the following sentence: These documents are a relevant part in the WYRED project, so we have planned to develop tools for showing and sharing them ### Assistant: VARIABLE: sharing them CONDITION: to develop CAUSE_1: These documents are a relevant part in the WYRED project KEYWORD: so we have planned AND: and EFFECT_1: to develop tools for showing EFFECT_2: sharing them ### END
### Instruction: Extract the cause and effect entities from the following sentence: The researchers are also an important part of the WYRED project, for this reason, they are going to have a private zone ### Assistant: KEYWORD: for this reason VARIABLE: they CONDITION: are also an important part of the WYRED project CAUSE_1: The researchers are also an important part of the WYRED project EFFECT_1: they are going to have a private zone ### END
### Instruction: Extract the cause and effect entities from the following sentence: In this zone, they will be able to monitor their own social dialogues, share information between the project researchers and analyze their data ### Assistant: KEYWORD: In VARIABLE: they CAUSE_1: this CONDITION: analyze their data AND: and EFFECT_1: they will be able to monitor their own social dialogues EFFECT_2: share information between the project researchers EFFECT_3: analyze their data ### END
### Instruction: Extract the cause and effect entities from the following sentence: In order to improve the data analysis process, we are going to develop visualization tools and a pattern matching system ### Assistant: KEYWORD: we are going CONDITION: to develop VARIABLE: a pattern matching system EFFECT_1: to improve the data analysis process AND: and CAUSE_1: to develop visualization tools CAUSE_2: a pattern matching system ### END
### Instruction: Extract the cause and effect entities from the following sentence: The pattern matching system will use techniques such as datamining or natural language processing, in order to give to the researchers some patterns about the stored data ### Assistant: KEYWORD: in order CONDITION: natural language processing VARIABLE: The pattern matching system EFFECT_1: to give to the researchers some patterns about the stored data CAUSE_1: The pattern matching system will use techniques such as datamining OR: or CAUSE_2: natural language processing ### END
### Instruction: Extract the cause and effect entities from the following sentence: The system shall use an automatic translation system such as Google translator in order to allow the communication between people who speaks different languages ### Assistant: KEYWORD: in order CONDITION: shall use an automatic translation system VARIABLE: The system EFFECT_1: to allow the communication between people who speaks different languages NOT_RELEVANT: such as Google translator CAUSE_1: The system shall use an automatic translation system ### END
### Instruction: Extract the cause and effect entities from the following sentence: The system shall have a system that allows annotation in video and audio files ### Assistant: KEYWORD: that allows CONDITION: shall have a system VARIABLE: The system AND: and EFFECT_1: annotation in video EFFECT_2: audio files CAUSE_1: The system shall have a system ### END
### Instruction: Extract the cause and effect entities from the following sentence: The system shall have monitoring tools in the researcher zone in order to see the social dialogue evolution ### Assistant: KEYWORD: in CONDITION: shall have monitoring tools VARIABLE: the researcher zone EFFECT_1: to see the social dialogue evolution CAUSE_1: The system shall have monitoring tools CAUSE_2: the researcher zone ### END
### Instruction: Extract the cause and effect entities from the following sentence: The planning will be updated following the assessment phase and the Contractor shall provide its own bottom-up development schedule ### Assistant: KEYWORD: following CONDITION: shall provide its own bottom-up development schedule CAUSE_1: the assessment phase VARIABLE: the Contractor EFFECT_1: The planning will be updated AND: and EFFECT_2: the Contractor shall provide its own bottom-up development schedule ### END
### Instruction: Extract the cause and effect entities from the following sentence: It will be reviewed as part of the assessment phase and will be updated following the Preliminary Requirement Review at the end of 2013 ### Assistant: KEYWORD: at CONDITION: will be updated CAUSE_1: the Preliminary Requirement Review VARIABLE: It CAUSE_2: the end of 2013 AND: and EFFECT_1: It will be reviewed as part of the assessment phase EFFECT_2: will be updated ### END
### Instruction: Extract the cause and effect entities from the following sentence: Any conflict between requirements in this document shall be reported to ESA and any modification requires his/her prior approval ### Assistant: CONDITION: shall be reported to ESA CAUSE_1: conflict between requirements in this document AND: and EFFECT_1: shall be reported to ESA KEYWORD: any CAUSE_2: modification VARIABLE: modification EFFECT_2: his/her prior approval ### END
### Instruction: Extract the cause and effect entities from the following sentence: They are highlighted in italic in order to be easily identified ### Assistant: KEYWORD: in order CONDITION: are highlighted in italic VARIABLE: They CAUSE_1: They are highlighted in italic EFFECT_1: to be easily identified ### END
### Instruction: Extract the cause and effect entities from the following sentence: Therefore the spacecraft performs a soft touchdown of the surface for a few seconds (in the order of 2-5 seconds) and then takes-off immediately after that to move into a safe position away from the surface ### Assistant: KEYWORD: then VARIABLE: the spacecraft CONDITION: takes-off immediately after that to move into a safe position away from the surface NOT_RELEVANT: (in the order of 2-5 seconds) AND: and CAUSE_1: the spacecraft performs a soft touchdown of the surface for a few seconds EFFECT_1: takes-off immediately after that to move into a safe position away from the surface ### END
### Instruction: Extract the cause and effect entities from the following sentence: If it is confirmed, via a reliable verification technique, that a scientifically meaningful sample has been collected, the sample is transferred to the re-entry capsule and sealed ### Assistant: KEYWORD: that VARIABLE: the sample CONDITION: sealed NOT_RELEVANT: via a reliable verification technique CAUSE_1: it is confirmed CAUSE_2: a scientifically meaningful sample has been collected AND: and EFFECT_1: the sample is transferred to the re-entry capsule EFFECT_2: sealed ### END
### Instruction: Extract the cause and effect entities from the following sentence: If the sampling operation was not successful, the spacecraft can undertake 2 more attempts (3 attempts in total) ### Assistant: KEYWORD: If VARIABLE: the spacecraft NEGATION: was not CONDITION: can undertake 2 more attempts CAUSE_1: the sampling operation was not successful NOT_RELEVANT: (3 attempts in total) EFFECT_1: the spacecraft can undertake 2 more attempts ### END
### Instruction: Extract the cause and effect entities from the following sentence: Any conflict between the requirements in this document and the ECSS standards shall be reported to ESA ### Assistant: KEYWORD: Any VARIABLE: conflict CONDITION: shall be reported to ESA CAUSE_1: conflict between the requirements in this document and the ECSS standards EFFECT_1: shall be reported to ESA ### END
### Instruction: Extract the cause and effect entities from the following sentence: Before peak of dynamic pressure, angle of attack variations are less critical ### Assistant: NEGATION: Before CONDITION: are less critical VARIABLE: angle of attack variations CAUSE_1: peak of dynamic pressure EFFECT_1: angle of attack variations are less critical ### END
### Instruction: Extract the cause and effect entities from the following sentence: The spacecraft design and operations shall transmit back to Earth a minimum total payload data volume of 122.35 GBit (TBC) before the end of the asteroid proximity operations ### Assistant: NEGATION: before CONDITION: shall transmit back to Earth a minimum total payload data volume of 122.35 GBit VARIABLE: operations CAUSE_1: the end of the asteroid proximity operations NOT_RELEVANT: (TBC) EFFECT_1: The spacecraft design AND: and EFFECT_2: operations shall transmit back to Earth a minimum total payload data volume of 122.35 GBit ### END
### Instruction: Extract the cause and effect entities from the following sentence: The spacecraft structure shall support the mechanical static and dynamic loads encountered during its entire lifetime, including: manufacturing, handling, transportation, testing, launch and in-orbit operations (incl ### Assistant: KEYWORD: during VARIABLE: The spacecraft structure CAUSE_1: its entire lifetime NOT_RELEVANT: including: manufacturing, handling, transportation, testing, launch and in-orbit operations (incl CONDITION: dynamic loads encountered AND: and EFFECT_1: The spacecraft structure shall support the mechanical static EFFECT_2: dynamic loads encountered ### END
### Instruction: Extract the cause and effect entities from the following sentence: R-AUT-030 During asteroid proximity operations, if a ground TC is expected for the S/C to proceed with its nominal manoeuvre and this TC is not received within a TBD time, the S/C shall be able to perform a valid transition into an asteroid Collision Avoidance Mode ### Assistant: KEYWORD: if AND: and NOT_RELEVANT: R-AUT-030 VARIABLE: the S/C CONDITION: shall be able to perform a valid transition into an asteroid Collision Avoidance Mode CAUSE_1: asteroid proximity operations CAUSE_2: a ground TC is expected for the S/C to proceed with its nominal manoeuvre NEGATION: is not CAUSE_3: this TC is not received within a TBD time EFFECT_1: the S/C shall be able to perform a valid transition into an asteroid Collision Avoidance Mode ### END
### Instruction: Extract the cause and effect entities from the following sentence: R-GRS-080 It shall be possible to track the ERC re-entry during the whole re-entry operation ### Assistant: KEYWORD: during VARIABLE: It CAUSE_1: the whole re-entry operation CONDITION: shall be possible to track the ERC re-entry NOT_RELEVANT: R-GRS-080 EFFECT_1: It shall be possible to track the ERC re-entry ### END
### Instruction: Extract the cause and effect entities from the following sentence: The ground segment shall provide for a flexible re-planning capability during the asteroid phase ### Assistant: KEYWORD: during VARIABLE: The ground segment CAUSE_1: the asteroid phase CONDITION: shall provide for a flexible re-planning capability EFFECT_1: The ground segment shall provide for a flexible re-planning capability ### END
### Instruction: Extract the cause and effect entities from the following sentence: The manoeuvre planning for new asteroid proximity orbits and new sampling site targets shall be available within 4 days (TBC) after reception of the respective request ### Assistant: NEGATION: after CONDITION: shall be available within 4 days VARIABLE: new sampling site targets CAUSE_1: reception of the respective request AND: and NOT_RELEVANT: (TBC) EFFECT_1: The manoeuvre planning for new asteroid proximity orbits EFFECT_2: new sampling site targets shall be available within 4 days ### END
### Instruction: Extract the cause and effect entities from the following sentence: The ground segment shall construct a global asteroid map and a global Digital Elevation Model with a 1-m spatial resolution (TBC) of the whole asteroid during the GCP phase primarily based on the navigation sensor information ### Assistant: KEYWORD: during VARIABLE: The ground segment CONDITION: primarily based on the navigation sensor information AND: and NOT_RELEVANT: (TBC) EFFECT_1: The ground segment shall construct a global asteroid map EFFECT_2: of the whole asteroid CAUSE_1: the GCP phase EFFECT_3: primarily based on the navigation sensor information ### END
### Instruction: Extract the cause and effect entities from the following sentence: The ground segment shall construct a local map and a local Digital Elevation Model with a 20-cm spatial resolution (TBC) of the sampling site candidates during the LCP phase primarily based on the navigation sensor information ### Assistant: VARIABLE: the LCP phase AND: and KEYWORD: during CONDITION: primarily based on the navigation sensor information NOT_RELEVANT: (TBC) EFFECT_1: The ground segment shall construct a local map EFFECT_2: of the sampling site candidates EFFECT_3: primarily based on the navigation sensor information CAUSE_1: the LCP phase ### END
### Instruction: Extract the cause and effect entities from the following sentence: Radio-science orbit: In the context of MarcoPolo-R, a Radio-science orbit is an asteroid-bound orbit, favourable to radio science measurements, where the S/C does not require any orbit and attitude manoeuvres in order to remain above 3 km distance to the surface at any time, during four consecutive communication windows of 8 hours ### Assistant: KEYWORD: In the context of CONDITION: favourable to radio science measurements VARIABLE: four consecutive communication windows NEGATION: does not AND: and CAUSE_1: MarcoPolo-R EFFECT_3: to remain above 3 km distance to the surface at any time CAUSE_3: four consecutive communication windows of 8 hours. EFFECT_1: a Radio-science orbit is an asteroid-bound orbit EFFECT_2: favourable to radio science measurements CAUSE_2: attitude manoeuvres NOT_RELEVANT: Radio-science orbit ### END
### Instruction: Extract the cause and effect entities from the following sentence: The minimum scientific success criteria for the mission are reached if the two core science objectives (TOP1 and TOP2) are achieved ### Assistant: KEYWORD: if VARIABLE: The minimum scientific success criteria for the mission CONDITION: are reached NOT_RELEVANT: (TOP1 and TOP2) EFFECT_1: The minimum scientific success criteria for the mission are reached CAUSE_1: are achieved ### END
### Instruction: Extract the cause and effect entities from the following sentence: After applying all known calibration factors, the linearity and offset of the energy scale applying to a measured spectrum shall allow a photon energy to be determined to within a small fraction ### Assistant: NEGATION: After CONDITION: to be determined to within a small fraction VARIABLE: a photon energy AND: and CAUSE_1: applying all known calibration factors EFFECT_1: the linearity EFFECT_2: offset of the energy scale KEYWORD: shall allow EFFECT_3: a photon energy to be determined to within a small fraction. CAUSE_2: applying to a measured spectrum ### END
### Instruction: Extract the cause and effect entities from the following sentence: Unstable response leads to apparent noise in constant flux sources and also to QPO broadening or sidebands ### Assistant: CONDITION: to apparent VARIABLE: QPO broadening KEYWORD: leads AND: and also to OR: or CAUSE_1: Unstable response EFFECT_1: to apparent noise in constant flux sources EFFECT_2: QPO broadening EFFECT_3: sidebands ### END
### Instruction: Extract the cause and effect entities from the following sentence: For pulsars, knowing the detection time with an accuracy of 2 microsecond allows comparison with other wavebands (e.g. radio) ### Assistant: KEYWORD: allows VARIABLE: comparison with other wavebands CONDITION: with an accuracy of 2 microsecond NOT_RELEVANT: (e.g. radio). CAUSE_1: pulsars CAUSE_2: knowing the detection time with an accuracy of 2 microsecond EFFECT_1: comparison with other wavebands ### END
### Instruction: Extract the cause and effect entities from the following sentence: Bright transient sources can be stored on-board and the full information can be transferred during subsequent ground contacts ### Assistant: KEYWORD: during CONDITION: can be stored on-board VARIABLE: the full information CAUSE_1: subsequent ground contacts AND: and EFFECT_1: Bright transient sources can be stored on-board EFFECT_2: the full information can be transferred ### END
### Instruction: Extract the cause and effect entities from the following sentence: Sensitivity (5 σ detection) over 50 ks must be < 5 mCrab over energy range of 2 - 50 keV for a field where there is no source confusion (outside the Galactic Centre) ### Assistant: KEYWORD: for NEGATION: is no CONDITION: must be < 5 mCrab over energy range of 2 - 50 keV CAUSE_2: a field NOT_RELEVANT: (5 σ detection) VARIABLE: a field CAUSE_1: over 50 ks EFFECT_1: must be < 5 mCrab over energy range of 2 - 50 keV ### END
### Instruction: Extract the cause and effect entities from the following sentence: If the LAD data rate is <80%, the WFM should be able to use up the available bandwidth ### Assistant: KEYWORD: If VARIABLE: the WFM CONDITION: should be able to use up the available bandwidth CAUSE_1: the LAD data rate is <80% EFFECT_1: the WFM should be able to use up the available bandwidth ### END
### Instruction: Extract the cause and effect entities from the following sentence: Energy range and telemetry bandwidth should allow at least 8 spectral bands for the detection of spectral state changes in transient sources and be sensitive to thermal (e.g. disk) spectral components ### Assistant: KEYWORD: for VARIABLE: telemetry bandwidth CONDITION: spectral components AND: and NOT_RELEVANT: (e.g. disk) CAUSE_1: the detection of spectral state changes in transient sources EFFECT_1: Energy range EFFECT_2: telemetry bandwidth should allow at least 8 spectral bands EFFECT_3: spectral components ### END
### Instruction: Extract the cause and effect entities from the following sentence: The transients may be detected both in triggered data and in normal data if the transient rise time is >100s or a spectral state change has occurred ### Assistant: KEYWORD: if VARIABLE: The transients CONDITION: in normal data OR: or CAUSE_1: the transient rise time is >100s CAUSE_2: a spectral state change has occurred AND: and EFFECT_1: The transients may be detected both in triggered data EFFECT_2: in normal data ### END
### Instruction: Extract the cause and effect entities from the following sentence: After a trigger into burst mode, the WFM will attempt to localise the source on the sky and determine the time of trigger ### Assistant: NEGATION: After CAUSE_1: a trigger into burst mode VARIABLE: a trigger CONDITION: into burst mode AND: and EFFECT_1: the WFM will attempt to localise the source on the sky EFFECT_2: determine the time of trigger ### END
### Instruction: Extract the cause and effect entities from the following sentence: Justification: If broadcasted, a set of ground receivers can see the trigger times and positions for follow up measurements ### Assistant: KEYWORD: If NOT_RELEVANT: Justification CONDITION: follow up CAUSE_1: broadcasted VARIABLE: measurements AND: and EFFECT_1: a set of ground receivers can see the trigger times EFFECT_2: positions CAUSE_2: follow up measurements ### END
### Instruction: Extract the cause and effect entities from the following sentence: Otherwise (50% sky visibility) the observation time when the bulge is visible to LOFT is used for a significant fraction by observing the core-science targets ### Assistant: KEYWORD: when VARIABLE: the observation time CONDITION: is used for a significant fraction by observing the core-science targets CAUSE_1: the bulge is visible to LOFT NOT_RELEVANT: (50% sky visibility) EFFECT_1: is used for a significant fraction by observing the core-science targets NEGATION: Otherwise ### END