Predicting learning outputs and retention through neural network artificial intelligence in photosynthesis, transpiration and translocation http://www.ied.edu.hk/apfslt/v17

Artificial Intelligence (AI) is a branch of computer science relates, connects, classifies, differentiates, and elaborates different cognitive domains in the neurological network. John Mc. Carthy coined the term in 1955 and viewed, “it is the science and engineering of making intelligent machines.” AI is highly technical, specialized, and based on deduction, reasoning, and problem solving techniques linked with mathematics, reasoning, logic, and engineering to identifying the concept in various situations. The central theme of AI research includes reasoning, planning, learning, communicating, perceiving, constructing knowledge (Jena, 2012), and manipulating the objects or concepts. This field includes human creativity, and intelligence to stimulate the cognition (Goyache, 2001). Recently, statistical methods and computational intelligence is included in AI for mathematical optimization and ontological classification while neural network stimulates human brain and intelligence, and in fact, the nature of mind stimulates to perform better or to give an appropriate response.

Now a day, online learning approaches are using to enhance science learning performance (Jena, 2014, 2015), but AI a new paradigm shift increases the levels of learners’ intelligence (Guastello & Rieke, 1994). Human intelligence is based on intuition, common sense, judgments, creativity, beliefs and ability to demonstrate the intelligence by communicating effectively through reasoning and critical thinking those stimulate human behaviour and cognitive process (Jolly et al. 2007). It needs human expertise for quick data entry, manipulation, accumulation, and assimilation of information to find a better result and long retention. AI is not a mixed knowledge but a developed knowledge while human intelligence is a mixed knowledge, is not able to write a huge amount of data in memory. In the 19th century, French scientist, and philosopher Claude Bernard argued, “we achieve more than we know, we know more than we understand and we understand more than we can explain”. AI sometimes contrasts with conventional computing process. It uses to search the matching code in the environment to decode the human intelligence but it’s just like the software process needs inputs, short term memory, and intervention for long term memory results to long term retention and high output while conventional computing system needs software, followed by logical series of input to reach a conclusion. The conventional computing system is a software process needs algorithms, but AI is an image interpretation labeling or a segmented image interprets the human intelligence for conceptualization of the knowledge and it also detects the unexpected conceptions, doubts and misconceptions (Khosrowshahi, 2011). Here, questions raised on how neural network does work effectively on learners’ achievement and retention. Does it help in detecting the misconception, if so, then how? Does the neural network encourage long-term retention, and how does it work to predict the output of the problems?

Science teachers are trying to teach the proposition or the conceptual network directly to stimulate the central nervous system of children, and many researchers have been conducting research on the effect of artificial intelligence on learner’s change of behaviour, and science ability of students (Jena, Gogoi, Deka, 2016). Shaw and Fox (1993) found that neural network could solve the learning difficulties or problem in different situation. However, game artificial intelligence provided reinforcement to the learners and speedup the learning process (Ajung & Gaol, 2012). Artificial intelligence as a technique helped learners to learn symbolic reasoning and increased the flexibility and capabilities of learning (Cavus, 2010; Mellit & Kalogirou, 2008).However, Conrad (1987) found the dynamic mechanism of artificial intelligence and it is a computer network topology (Pierre, 1993). Not only was that, Conrad (1987) found that artificial intelligence is a dynamic mechanism of computer network topology (Pierre, 1993). Artificial intelligence demonstrates the feasibility of the approach and found 50% more improvement among the student (Bennett & Hauser, 2013). In addition to that, it was found, artificial neural network is a learning environment use to embedded algorithm to enhance the quality of solution, and points out differences in performance between light and standard bovine carcasses (Bahamond et al, 2003; Noroozi et al, 2013). Estimating environmental pollution through Artificial Intelligence is successful (Compare, 1998) as comparable to other techniques. Machine learning in Artificial Intelligence uses to teach different concepts through inductive and logical programming (Bratko, 1993).

Not only was that but also the prediction of cotton yarn properties was assessed through artificial intelligence was successful (Stjepanovic & Jezernik, 1991). However, neural network and machine learning are the kind of artificial intelligence (Prieto et al, 2013). Artificial Intelligence is a machine learning technique manages complexity, changes and uncertainties and can predicts productivity (Barto & Sutton, 1997, Hendry, 1987, Monostory, 2013) and inspired social intelligence (Dautenhahn, 1995). Artificial neural network intelligence is an approach of induction vs. self-organizing neural networks (Mullohand, 1995) could restructure the knowledge and it is an efficient technique for predicting the learning out comes (Kolodziejezyk, 2010) especially, used as the semantic analyzer (Feldman & Yakimovsky, 1974). Literatures, found that artificial intelligence effects on mind, and achievement and it is an effective approach for teaching and learning process could predict learning outputs. That’s why the study aimed 1) to study the existing misconception of students in science learning; 2) to study the effectiveness of neural network artificial intelligence approach on the prediction of learners’ learning outputs; and 3) to study the effectiveness of neural network artificial intelligence approach on the achievement and retention of science learning. Based on the objectives the study assumed that 1) there are misconceptions of students in science, 2) there are significant effects of neural network artificial intelligence approach on the prediction of learners’ learning outputs, and 3) there are significant effects of neural network artificial intelligence approach on achievement and retention of science learning.