The title should clearly inform the reader what to expect in the paper. Avoid common descriptors, such as "novel" or "performance evaluation." Use words that zero in on the distinct features of your study -- words people interested in your research would likely enter into a search engine. The title is the reader's first decision point; he or she may choose to read further or simply move on.
The abstract must concisely explain the nature and purpose of your research and why anyone would be interested in reading about it. The abstract may cite the work of different authors in the text, but should not contain references, as it is meant to be a stand-alone piece of text. Highlight the purpose of the research and its main results. As in the title, be sure to include the key terms most likely to be used by search engines to find your work.
The introduction should clearly and immediately tell the reader what the paper is about. Get to the point, and get to it quickly. A reader will not read more than a few paragraphs before deciding whether reading the rest is worthwhile. A good introduction provides a preview of the rest of the paper. It sets up the problem and provides an overview of your approach, solution and results. Do not repeat the abstract in the introduction.
This section provides an overview of the materials and methods used in your research. Another researcher should be able to read this section to assess the merits of your study or to use the materials and methods described in a similar study. Include in this section specialized materials and equipment, not standard lab equipment such as test tubes and beakers. When describing your methodology, do not provide step-by-step instructions. Report how procedures were generally done, not how they were specifically executed in one instance. If you used a well-documented methodology, refer to it by name.
Present and illustrate your findings in a concise, objective manner. Save all interpretation of results for the discussion section. Summarize your results in writing and highlight them with illustrations, tables and figures. Provide context for your results -- describing how a particular observation answered a particular question -- and direct your reader to the most relevant results. Do not include raw data.
This is your chance to interpret your results, supporting all of your conclusions with evidence from your research. If your results differed from your expectations, explain what happened. If your results support your thesis, elaborate on the significance of your findings. Do not dismiss a study as "inconclusive," but draw conclusions based on the results you have and suggest further research. Recommend how to improve upon similar research in the future.
Cite all references in your bibliography, listing the source, date and other identifying information. Avoid using "et al." in a bibliography unless the source lists five or more authors. Include publication years in book citations, but no ISBNs. You may include URLs to reference software and other non-library materials. Mark Internet drafts "work in progress," and look for the newest versions or RFCs (Request for Comments). For conference papers, list the conference location, month and full conference name.
Acknowledge your funding sources, and check if your source has specific wording requirements. You do not have to acknowledge anonymous reviewers, unless they provided an extraordinary amount of help.